v1.15.0-beta.3

1.15.0

• Ci(android): Play Store alpha uploads to the internal track now publish as completed so testers can download without a Console roll-out

• Fix(sync): Reject WebDAV database upload when the local library has no non-deleted books, to avoid wiping the cloud library with an empty DB (#911, #898 mitigation)

• Feat(notes): Export/import book notes as JSON with book fingerprint (md5/title/author) for per-book migration across devices (#898, #911 escape hatch)

• Fix(webdav): Follow 301/307/308 redirects (vendored webdav_client) so AList-style redirect strategies work (#345)

• Fix(android): Optional setting to trust user-installed / custom CA certificates for HTTPS WebDAV (e.g. private Nextcloud) (#719)

• Feat(sync): WebDAV URL hint — fill a writable directory; the app creates anx/ under it

• Feat(android): Long-press expands CJK selection to a word (Intl.Segmenter); selection handles remain freely draggable

• Fix(reader): Honor writing-direction setting for EPUBs that set writing-mode on body (e.g. vertical Japanese books) (#867)

• Fix(log): Silence Chromium iframe sandbox WebView warning that already had an ignore entry but failed exact match due to trailing period (#877)

• Ci: Upgrade lock-threads to v6 and use github.token so lock-closed-issues stops failing daily on long GITHUB_TOKEN secrets

• Feat(search): Add in-app search results with configurable search engine and display modes (#894)

• Feat(ai): Add configurable delay threshold for auto-summary when reopening a book (#922)

• Feat(ai): Make custom user prompts available in the Home AI tab for parity with reader AI (#853)

• Fix(bookshelf): Show book file format in detail metadata so users can see the extension inside the app (#910)

• Feat(network): Add global HTTP proxy support in advanced settings (#838) Thanks @dddXzz

• Feat(network): Add HTTP proxy connectivity test feature (#838) Thanks @dddXzz

• Fix(reader): Fix Android selection auto-page turn — continuous turns and missed turns when dragging across pages (#875) Thanks @addtion99

• Fix(l10n): Update Russian translation (#874) Thanks @Xapitonov

• Fix(ai): Avoid native WebView2 crash when running book content search on Windows (#978) Thanks @bazzdug-arch

• Perf(reader): Reduce Android reading-page scroll jank by deferring relocate during scroll (#914) Thanks @dddXzz

• Fix(reader): Restore reader keyboard focus after clearing text selection / closing context menu (#967, #966) Thanks @zyx-31415

• Fix(reader): Open selection menu on single-word long-press on Android (#990, #968, #900)

• Fix(sync): Strip special characters (e.g. # @ $ %) from file names on import so WebDAV servers like Jianguoyun accept uploads (#989)

• Feat(appearance): Disable open-book animation, page transitions, and dialog motion when e-ink mode is on (#986)

• Fix(bookshelf): Make group rename more obvious with an edit icon (#972)

• Fix(ai): Preserve Gemini thought_signature across tool-call round-trips so thinking models work with tools (#977)

• Fix(reader): Page-turn keys clear selection and turn pages instead of panning while text is selected (#966)

• Fix(ai): Null-safe custom provider config parsing and clearer errors when required fields or response shape are invalid (#868)

• Fix(ai): Quick prompt chips insert into the composer for editing; long-press still sends immediately (#969)

• Feat(linux): Add Linux AppImage and tar.gz packaging via CEF (#952)

• Fix(ai): Make the Add Prompt button in AI settings full-width to match the surrounding rows

• Fix(ai): Fix AI settings showing a grey error screen on fresh installs and provider edits failing to save, by restoring generated AiProvider JSON serialization with lenient converters

• Chore(android): Target Android 16 (API 36) and upgrade to Google Play Billing Library 8 to meet Google Play requirements

• Ci(android): Play 商店 alpha 上传至 internal 轨道时自动设为 completed，测试者无需在 Console 手动发布即可下载

• Fix(sync): 本地书库无未删除书籍时拒绝上传 WebDAV 数据库，避免用空库覆盖云端 (#911, #898 缓解)

• Feat(notes): 支持按书导出/导入笔记 JSON（含 md5/书名/作者指纹），便于跨设备迁移批注 (#898, #911 逃生舱)

• Fix(webdav): 跟随 301/307/308 重定向（vendored webdav_client），兼容 AList 等 302 策略 (#345)

• Fix(android): 新增可选设置以信任用户安装/自定义 CA 证书，便于私有 Nextcloud 等 HTTPS WebDAV (#719)

• Feat(sync): WebDAV URL 提示——填写可写入目录，应用会在其下创建 anx/

• Feat(android): 长按单字 CJK 选区通过 Intl.Segmenter 扩展为词语，仍可自由拖动手柄调整选区

• Fix(reader): 修复部分 EPUB（如在 body 上设置竖排 writing-mode）忽略写作方向设置的问题 (#867)

• Fix(log): 修复因句末句点导致精确匹配失败、未能屏蔽 Chromium iframe sandbox WebView 警告的问题 (#877)

• Ci: 升级 lock-threads 至 v6 并改用 github.token，避免 lock-closed-issues 因过长 GITHUB_TOKEN 每日失败

• Feat(search): 支持应用内搜索结果显示，可配置搜索引擎和显示方式 (#894)

• Feat(ai): 新增自动摘要延迟阈值配置，可控制重新打开书籍时的摘要触发时机 (#922)

• Feat(ai): 让自定义用户提示词在首页 AI 标签页中可用，并与阅读器 AI 保持一致 (#853)

• Fix(bookshelf): 在书籍详情元数据中显示文件格式，方便用户在应用内查看扩展名 (#910)

• Feat(network): 新增全局 HTTP 代理支持，可在高级设置中配置代理服务器地址和端口 (#838) 感谢 @dddXzz

• Feat(network): 新增代理连接测试功能 (#838) 感谢 @dddXzz

• Fix(reader): 修复 Android 选区跨页自动翻页的连续翻页与漏翻问题 (#875) 感谢 @addtion99

• Fix(l10n): 更新俄语翻译 (#874) 感谢 @Xapitonov

• Fix(ai): 修复 Windows 上书籍内容搜索触发原生 WebView2 崩溃的问题 (#978) 感谢 @bazzdug-arch

• Perf(reader): 优化 Android 阅读页滚动卡顿，滚动期间推迟 relocate (#914) 感谢 @dddXzz

• Fix(reader): 修复清除选区/关闭上下文菜单后阅读器键盘焦点未恢复的问题 (#967, #966) 感谢 @zyx-31415

• Fix(reader): 修复 Android 长按选中单字不弹出菜单的问题 (#990, #968, #900)

• Fix(sync): 导入时移除文件名中的特殊字符（如 # @ $ %），避免坚果云等 WebDAV 拒绝上传 (#989)

• Feat(appearance): 开启墨水屏模式时关闭开书动画、页面转场与弹窗动画 (#986)

• Fix(bookshelf): 分组重命名入口增加编辑图标，更加明显 (#972)

• Fix(ai): 修复 Gemini 工具调用时缺失 thought_signature 导致思考模型报错的问题 (#977)

• Fix(reader): 选中文本时方向键/翻页键先清除选区再翻页，避免页面平移 (#966)

• Fix(ai): 自定义供应商配置解析更空安全，并在必填项或响应格式异常时给出明确错误 (#868)

• Fix(ai): 快捷提示词芯片点击写入输入框以便编辑，长按仍可立即发送 (#969)

• Feat(linux): 新增 Linux AppImage 与 tar.gz 打包支持（基于 CEF）(#952)

• Fix(ai): AI 设置中“添加提示词”按钮改为占满整行，与上下设置项对齐

• Fix(ai): 修复全新安装时 AI 设置页灰屏、供应商增删改无法保存的问题（恢复 AiProvider 生成式 JSON 序列化并保留宽松解析）

• Chore(android): 目标 SDK 升级到 Android 16（API 36），并升级到 Google Play Billing Library 8，满足 Google Play 上架要求

v1.15.0-beta.2

1.15.0

• Fix(reader): Honor writing-direction setting for EPUBs that set writing-mode on body (e.g. vertical Japanese books) (#867)

• Fix(log): Silence Chromium iframe sandbox WebView warning that already had an ignore entry but failed exact match due to trailing period (#877)

• Ci: Upgrade lock-threads to v6 and use github.token so lock-closed-issues stops failing daily on long GITHUB_TOKEN secrets

• Feat(search): Add in-app search results with configurable search engine and display modes (#894)

• Feat(ai): Add configurable delay threshold for auto-summary when reopening a book (#922)

• Feat(ai): Make custom user prompts available in the Home AI tab for parity with reader AI (#853)

• Fix(bookshelf): Show book file format in detail metadata so users can see the extension inside the app (#910)

• Feat(network): Add global HTTP proxy support in advanced settings (#838) Thanks @dddXzz

• Feat(network): Add HTTP proxy connectivity test feature (#838) Thanks @dddXzz

• Fix(reader): Fix Android selection auto-page turn — continuous turns and missed turns when dragging across pages (#875) Thanks @addtion99

• Fix(l10n): Update Russian translation (#874) Thanks @Xapitonov

• Fix(ai): Avoid native WebView2 crash when running book content search on Windows (#978) Thanks @bazzdug-arch

• Perf(reader): Reduce Android reading-page scroll jank by deferring relocate during scroll (#914) Thanks @dddXzz

• Fix(reader): Restore reader keyboard focus after clearing text selection / closing context menu (#967, #966) Thanks @zyx-31415

• Fix(reader): Open selection menu on single-word long-press on Android (#990, #968, #900)

• Fix(sync): Strip special characters (e.g. # @ $ %) from file names on import so WebDAV servers like Jianguoyun accept uploads (#989)

• Feat(appearance): Disable open-book animation, page transitions, and dialog motion when e-ink mode is on (#986)

• Fix(bookshelf): Make group rename more obvious with an edit icon (#972)

• Fix(ai): Preserve Gemini thought_signature across tool-call round-trips so thinking models work with tools (#977)

• Fix(reader): Page-turn keys clear selection and turn pages instead of panning while text is selected (#966)

• Fix(ai): Null-safe custom provider config parsing and clearer errors when required fields or response shape are invalid (#868)

• Fix(ai): Quick prompt chips insert into the composer for editing; long-press still sends immediately (#969)

• Feat(linux): Add Linux AppImage and tar.gz packaging via CEF (#952)

• Fix(ai): Make the Add Prompt button in AI settings full-width to match the surrounding rows

• Fix(ai): Fix AI settings showing a grey error screen on fresh installs and provider edits failing to save, by restoring generated AiProvider JSON serialization with lenient converters

• Chore(android): Target Android 16 (API 36) and upgrade to Google Play Billing Library 8 to meet Google Play requirements

• Fix(reader): 修复部分 EPUB（如在 body 上设置竖排 writing-mode）忽略写作方向设置的问题 (#867)

• Fix(log): 修复因句末句点导致精确匹配失败、未能屏蔽 Chromium iframe sandbox WebView 警告的问题 (#877)

• Ci: 升级 lock-threads 至 v6 并改用 github.token，避免 lock-closed-issues 因过长 GITHUB_TOKEN 每日失败

• Feat(search): 支持应用内搜索结果显示，可配置搜索引擎和显示方式 (#894)

• Feat(ai): 新增自动摘要延迟阈值配置，可控制重新打开书籍时的摘要触发时机 (#922)

• Feat(ai): 让自定义用户提示词在首页 AI 标签页中可用，并与阅读器 AI 保持一致 (#853)

• Fix(bookshelf): 在书籍详情元数据中显示文件格式，方便用户在应用内查看扩展名 (#910)

• Feat(network): 新增全局 HTTP 代理支持，可在高级设置中配置代理服务器地址和端口 (#838) 感谢 @dddXzz

• Feat(network): 新增代理连接测试功能 (#838) 感谢 @dddXzz

• Fix(reader): 修复 Android 选区跨页自动翻页的连续翻页与漏翻问题 (#875) 感谢 @addtion99

• Fix(l10n): 更新俄语翻译 (#874) 感谢 @Xapitonov

• Fix(ai): 修复 Windows 上书籍内容搜索触发原生 WebView2 崩溃的问题 (#978) 感谢 @bazzdug-arch

• Perf(reader): 优化 Android 阅读页滚动卡顿，滚动期间推迟 relocate (#914) 感谢 @dddXzz

• Fix(reader): 修复清除选区/关闭上下文菜单后阅读器键盘焦点未恢复的问题 (#967, #966) 感谢 @zyx-31415

• Fix(reader): 修复 Android 长按选中单字不弹出菜单的问题 (#990, #968, #900)

• Fix(sync): 导入时移除文件名中的特殊字符（如 # @ $ %），避免坚果云等 WebDAV 拒绝上传 (#989)

• Feat(appearance): 开启墨水屏模式时关闭开书动画、页面转场与弹窗动画 (#986)

• Fix(bookshelf): 分组重命名入口增加编辑图标，更加明显 (#972)

• Fix(ai): 修复 Gemini 工具调用时缺失 thought_signature 导致思考模型报错的问题 (#977)

• Fix(reader): 选中文本时方向键/翻页键先清除选区再翻页，避免页面平移 (#966)

• Fix(ai): 自定义供应商配置解析更空安全，并在必填项或响应格式异常时给出明确错误 (#868)

• Fix(ai): 快捷提示词芯片点击写入输入框以便编辑，长按仍可立即发送 (#969)

• Feat(linux): 新增 Linux AppImage 与 tar.gz 打包支持（基于 CEF）(#952)

• Fix(ai): AI 设置中“添加提示词”按钮改为占满整行，与上下设置项对齐

• Fix(ai): 修复全新安装时 AI 设置页灰屏、供应商增删改无法保存的问题（恢复 AiProvider 生成式 JSON 序列化并保留宽松解析）

• Chore(android): 目标 SDK 升级到 Android 16（API 36），并升级到 Google Play Billing Library 8，满足 Google Play 上架要求

v1.15.0-alpha.9

Alpha build from staging.

Included staging updates:

• #936 feat(ai): show user prompts on home AI tab
• #937 fix(bookshelf): show book file format in details

v1.15.0-beta.1

1.15.0

• Feat(network): Add global HTTP proxy support in advanced settings (#838) Thanks @dddXzz

• Feat(network): Add HTTP proxy connectivity test feature (#838) Thanks @dddXzz

• Fix(reader): Fix Android selection auto-page turn — continuous turns and missed turns when dragging across pages (#875) Thanks @addtion99

• Fix(l10n): Update Russian translation (#874) Thanks @Xapitonov

• Feat(network): 新增全局 HTTP 代理支持，可在高级设置中配置代理服务器地址和端口 (#838) 感谢 @dddXzz

• Feat(network): 新增代理连接测试功能 (#838) 感谢 @dddXzz

• Fix(reader): 修复 Android 选区跨页自动翻页的连续翻页与漏翻问题 (#875) 感谢 @addtion99

• Fix(l10n): 更新俄语翻译 (#874) 感谢 @Xapitonov

v1.14.0

1.14.0

• Fix(translate): Remove legacy Microsoft reverse-engineered translation service and migrate saved full-text translation preference to Microsoft Azure API

• Fix(l10n): Remove legacy Microsoft translation localization entries

• Feat(ai): Support separate AI reasoning content with lightweight collapsible thinking UI in chat and stream views (#787)

• Feat(reader): Add background image fit mode setting with cover and stretch options

• Feat(appearance): Add setting to toggle action button labels visibility in selection context menu

• Feat(reader): Add background image blur and opacity controls in reading settings (#753)

• Feat(tts): Add Narrator option to text selection toolbar to start TTS from selected text (#794) Thanks @deskangel

• Feat(tts): Add click to pause/resume TTS playback on currently reading text (#794) Thanks @deskangel

• Feat(tts): Add floating action button in reader for quick TTS controls (previous, pause/resume, next, stop) (#723)

• Feat(reader): Add Ctrl+[ and Ctrl+] page turning shortcuts on macOS, support Logitech Options+ mouse button mapping (#794) Thanks @deskangel

• Feat(ai): Add global AI RPM rate limiting in AI service layer

• Feat(ai): Add AI Provider Configuration Center with support for OpenAI-compatible, Claude, and Gemini protocols

• Feat(ai): Support multiple API keys per provider with round-robin rotation

• Feat(ai): Add provider-level reasoning_effort configuration with auto mode and advanced settings entry

• Feat(ai): Add provider test connection with streaming preview

• Feat(ai): Add AI chat display mode settings with adaptive, split, and popup options

• Feat(ai): Add resizable AI panel with drag to resize, sizes persisted

• Feat(ai): Add AI panel position settings (bottom/right) for non-popup modes

• Feat(ai): Add font size setting for AI chat via three-dot menu

• Feat(ai): Add quick model switcher in AI chat input bar (tune icon button)

• Feat(ai): Add model picker in provider detail page with showMenu dropdown after fetching

• Fix(ui): Fix context menu action buttons layout and icon color issues when labels are hidden

• Fix(reader): Fix background image effects not applying and switch reader background fill mode to cover (#753)

• Fix(reader): Fix RangeError crash when read theme color is null/invalid (#759)

• Fix(tts): Fix TTS type selection having no effect when changed in reading interface (#794) Thanks @deskangel

• Fix(tts): Fix incorrect reading position after modifying pitch or rate (#794) Thanks @deskangel

• Fix(tts): Fix crash when SystemTts speak() receives null text from WebView

• Fix(ai): Fix unable to use AI for full-text translation

• Fix(ai): Fix test connection using default provider instead of current provider

• Fix(ai): Fix AI crash prevention by adding guards for null webViewEnvironment on Windows

• Fix(ai): Fix type cast error when using Gemini AI with tools (#747)

• Fix(reader): Desktop resource lifecycle, WebView stability, and scroll UX optimization (#790) Thanks @yi124773651

• Fix(reader): Fix window close cleanup to properly stop server, dispose WebView2, close database, and destroy window

• Fix(reader): Add scroll debounce mechanism for smoother page turning experience

• Fix(reader): Fix image saving permission issue on Android 10+ devices by removing unnecessary storage permission requests (#793)

• Fix(reader): Disable WebView2 right-click context menu (back, reload, save as, print) on Windows (#746)

• Fix(reader): Fix images with large inline max-width overflowing column width and appearing on wrong page (#742)

• Fix(reader): Fix epub heading color not following theme color when useBookStyles is disabled (#749)

• Fix(toc): Show full chapter titles in TOC by wrapping long text instead of truncating with ellipsis (#751)

• Fix(l10n): Fallback to English when system language is unsupported on first launch (#785)

• Feat(reader): Support independent left/right margins and font size for header and footer reading info, with shared section-based configuration model (#791)

• Fix(l10n): Update Russian translation in app_ru.arb (#774) Thanks @Xapitonov

• Fix(l10n): Update full_description.txt (#775) Thanks @Xapitonov

• L10n: Complete Romanian translation (#714) Thanks @Steinhagen

• Feat(ai): AI 支持单独的思考内容字段，并在聊天与流式界面中以轻量可折叠方式展示思考过程 (#787)

• Fix(translate): 移除旧版 Microsoft 逆向翻译服务，并将已保存的全文翻译偏好迁移到 Microsoft Azure API

• Fix(l10n): 移除旧版 Microsoft 翻译相关本地化条目

• Feat(appearance): 新增设置项，可开关选中文本菜单等操作按钮的底部文字提示

• Feat(reader): 新增背景图缩放模式设置，支持 cover 和 stretch 两种模式

• Feat(reader): 阅读设置新增背景图模糊度和透明度调节 (#753)

• Feat(tts): 在文本选择工具栏新增朗读选项，支持从选中文本开始 TTS (#794) 感谢 @deskangel

• Feat(tts): 新增点击正在朗读的文本以暂停/继续 TTS 播放 (#794) 感谢 @deskangel

• Feat(tts): 新增阅读界面朗读快捷操作悬浮按钮（上一句、暂停/继续、下一句、停止）(#723)

• Feat(reader): macOS 新增 Ctrl+[ 和 Ctrl+] 翻页快捷键，支持 Logitech Options+ 鼠标按键映射 (#794) 感谢 @deskangel

• Feat(ai): 在 AI 服务层添加全局 AI 请求限流（RPM）

• Feat(ai): 新增 AI 供应商配置中心，支持 OpenAI 兼容、Claude 和 Gemini 协议

• Feat(ai): 支持每个供应商配置多个 API Key，自动轮询使用

• Feat(ai): 新增供应商级 reasoning_effort 配置，支持自动模式并收纳到高级设置中

• Feat(ai): 新增供应商连接测试，支持流式预览

• Feat(ai): 新增 AI 对话显示模式设置，支持自适应、分屏和弹出三种模式

• Feat(ai): 新增可调整尺寸的 AI 面板，支持拖动调整大小，尺寸自动保存

• Feat(ai): 新增 AI 面板位置设置（非弹出模式下可选择底部或右侧）

• Feat(ai): 新增 AI 对话字体大小设置，通过三点菜单访问

• Feat(ai): 新增 AI 聊天输入栏快速切换模型按钮（调谐图标）

• Feat(ai): 供应商详情页新增模型选择器，获取后通过下拉菜单显示

• Fix(ui): 修复关闭文字提示时上下文菜单按钮布局及图标颜色变灰的问题

• Fix(reader): 修复背景图效果不生效的问题，并将阅读器背景图填充方式调整为 cover (#753)

• Fix(reader): 修复阅读主题颜色为空/无效时的 RangeError 崩溃 (#759)

• Fix(tts): 修复在阅读界面切换 TTS 类型无效的问题 (#794) 感谢 @deskangel

• Fix(tts): 修复调整音调或语速后朗读位置不正确的问题 (#794) 感谢 @deskangel

• Fix(tts): 修复 SystemTts 从 WebView 获取文本为空时的崩溃问题

• Fix(ai): 修复无法使用 AI 进行全文翻译的问题

• Fix(ai): 修复测试连接时使用默认供应商而非当前供应商的问题

• Fix(ai): 修复 AI 崩溃问题，为 Windows 上空 webViewEnvironment 添加防护

• Fix(ai): 修复使用 Gemini AI 工具时的类型转换错误 (#747)

• Fix(reader): 桌面端资源生命周期管理、WebView 稳定性及滚动体验优化 (#790) 感谢 @yi124773651

• Fix(reader): 修复窗口关闭时的清理流程，正确停止服务器、释放 WebView2、关闭数据库并销毁窗口

• Fix(reader): 添加滚动防抖机制，优化翻页体验

• Fix(reader): 修复 Android 10+ 设备图片保存权限问题，移除不必要的存储权限请求 (#793)

• Fix(reader): 修复 Windows 端右键触发 WebView2 系统上下文菜单（返回、刷新、另存为、打印）的问题 (#746)

• Fix(reader): 修复图片内嵌大 max-width 导致宽度超出栏宽而显示到下一页的问题 (#742)

• Fix(reader): 修复禁用书籍自带样式时 epub 标题颜色不跟随主题色的问题，智能识别黑色标题并进行覆盖 (#749)

• Fix(toc): 目录中过长的章节标题现在会自动换行显示完整内容，不再以省略号截断 (#751)

• Fix(l10n): 修复首次启动时系统语言不受支持会错误回退到阿拉伯语的问题，改为回退到英文 (#785)

• Feat(reader): 页眉和页脚阅读信息支持分别设置左右边距与字体大小，并重构为共享的分区配置模型 (#791)

• Fix(l10n): 更新俄语翻译 app_ru.arb (#774) 感谢 @Xapitonov

• L10n: 完成罗马尼亚语翻译 (#714) 感谢 @Steinhagen

v1.14.0-beta.3

1.14.0

• Feat(ai): Support separate AI reasoning content with lightweight collapsible thinking UI in chat and stream views (#787)

• Feat(reader): Add background image fit mode setting with cover and stretch options

• Feat(appearance): Add setting to toggle action button labels visibility in selection context menu

• Feat(reader): Add background image blur and opacity controls in reading settings (#753)

• Feat(tts): Add Narrator option to text selection toolbar to start TTS from selected text (#794) Thanks @deskangel

• Feat(tts): Add click to pause/resume TTS playback on currently reading text (#794) Thanks @deskangel

• Feat(tts): Add floating action button in reader for quick TTS controls (previous, pause/resume, next, stop) (#723)

• Feat(reader): Add Ctrl+[ and Ctrl+] page turning shortcuts on macOS, support Logitech Options+ mouse button mapping (#794) Thanks @deskangel

• Feat(ai): Add global AI RPM rate limiting in AI service layer

• Feat(ai): Add AI Provider Configuration Center with support for OpenAI-compatible, Claude, and Gemini protocols

• Feat(ai): Support multiple API keys per provider with round-robin rotation

• Feat(ai): Add provider-level reasoning_effort configuration with auto mode and advanced settings entry

• Feat(ai): Add provider test connection with streaming preview

• Feat(ai): Add AI chat display mode settings with adaptive, split, and popup options

• Feat(ai): Add resizable AI panel with drag to resize, sizes persisted

• Feat(ai): Add AI panel position settings (bottom/right) for non-popup modes

• Feat(ai): Add font size setting for AI chat via three-dot menu

• Feat(ai): Add quick model switcher in AI chat input bar (tune icon button)

• Feat(ai): Add model picker in provider detail page with showMenu dropdown after fetching

• Fix(ui): Fix context menu action buttons layout and icon color issues when labels are hidden

• Fix(reader): Fix background image effects not applying and switch reader background fill mode to cover (#753)

• Fix(reader): Fix RangeError crash when read theme color is null/invalid (#759)

• Fix(tts): Fix TTS type selection having no effect when changed in reading interface (#794) Thanks @deskangel

• Fix(tts): Fix incorrect reading position after modifying pitch or rate (#794) Thanks @deskangel

• Fix(tts): Fix crash when SystemTts speak() receives null text from WebView

• Fix(ai): Fix unable to use AI for full-text translation

• Fix(ai): Fix test connection using default provider instead of current provider

• Fix(ai): Fix AI crash prevention by adding guards for null webViewEnvironment on Windows

• Fix(ai): Fix type cast error when using Gemini AI with tools (#747)

• Fix(reader): Desktop resource lifecycle, WebView stability, and scroll UX optimization (#790) Thanks @yi124773651

• Fix(reader): Fix window close cleanup to properly stop server, dispose WebView2, close database, and destroy window

• Fix(reader): Add scroll debounce mechanism for smoother page turning experience

• Fix(reader): Fix image saving permission issue on Android 10+ devices by removing unnecessary storage permission requests (#793)

• Fix(reader): Disable WebView2 right-click context menu (back, reload, save as, print) on Windows (#746)

• Fix(reader): Fix images with large inline max-width overflowing column width and appearing on wrong page (#742)

• Fix(reader): Fix epub heading color not following theme color when useBookStyles is disabled (#749)

• Fix(toc): Show full chapter titles in TOC by wrapping long text instead of truncating with ellipsis (#751)

• Fix(l10n): Fallback to English when system language is unsupported on first launch (#785)

• Feat(reader): Support independent left/right margins and font size for header and footer reading info, with shared section-based configuration model (#791)

• Fix(l10n): Update Russian translation in app_ru.arb (#774) Thanks @Xapitonov

• Fix(l10n): Update full_description.txt (#775) Thanks @Xapitonov

• L10n: Complete Romanian translation (#714) Thanks @Steinhagen

• Feat(ai): AI 支持单独的思考内容字段，并在聊天与流式界面中以轻量可折叠方式展示思考过程 (#787)

• Feat(appearance): 新增设置项，可开关选中文本菜单等操作按钮的底部文字提示

• Feat(reader): 新增背景图缩放模式设置，支持 cover 和 stretch 两种模式

• Feat(reader): 阅读设置新增背景图模糊度和透明度调节 (#753)

• Feat(tts): 在文本选择工具栏新增朗读选项，支持从选中文本开始 TTS (#794) 感谢 @deskangel

• Feat(tts): 新增点击正在朗读的文本以暂停/继续 TTS 播放 (#794) 感谢 @deskangel

• Feat(tts): 新增阅读界面朗读快捷操作悬浮按钮（上一句、暂停/继续、下一句、停止）(#723)

• Feat(reader): macOS 新增 Ctrl+[ 和 Ctrl+] 翻页快捷键，支持 Logitech Options+ 鼠标按键映射 (#794) 感谢 @deskangel

• Feat(ai): 在 AI 服务层添加全局 AI 请求限流（RPM）

• Feat(ai): 新增 AI 供应商配置中心，支持 OpenAI 兼容、Claude 和 Gemini 协议

• Feat(ai): 支持每个供应商配置多个 API Key，自动轮询使用

• Feat(ai): 新增供应商级 reasoning_effort 配置，支持自动模式并收纳到高级设置中

• Feat(ai): 新增供应商连接测试，支持流式预览

• Feat(ai): 新增 AI 对话显示模式设置，支持自适应、分屏和弹出三种模式

• Feat(ai): 新增可调整尺寸的 AI 面板，支持拖动调整大小，尺寸自动保存

• Feat(ai): 新增 AI 面板位置设置（非弹出模式下可选择底部或右侧）

• Feat(ai): 新增 AI 对话字体大小设置，通过三点菜单访问

• Feat(ai): 新增 AI 聊天输入栏快速切换模型按钮（调谐图标）

• Feat(ai): 供应商详情页新增模型选择器，获取后通过下拉菜单显示

• Fix(ui): 修复关闭文字提示时上下文菜单按钮布局及图标颜色变灰的问题

• Fix(reader): 修复背景图效果不生效的问题，并将阅读器背景图填充方式调整为 cover (#753)

• Fix(reader): 修复阅读主题颜色为空/无效时的 RangeError 崩溃 (#759)

• Fix(tts): 修复在阅读界面切换 TTS 类型无效的问题 (#794) 感谢 @deskangel

• Fix(tts): 修复调整音调或语速后朗读位置不正确的问题 (#794) 感谢 @deskangel

• Fix(tts): 修复 SystemTts 从 WebView 获取文本为空时的崩溃问题

• Fix(ai): 修复无法使用 AI 进行全文翻译的问题

• Fix(ai): 修复测试连接时使用默认供应商而非当前供应商的问题

• Fix(ai): 修复 AI 崩溃问题，为 Windows 上空 webViewEnvironment 添加防护

• Fix(ai): 修复使用 Gemini AI 工具时的类型转换错误 (#747)

• Fix(reader): 桌面端资源生命周期管理、WebView 稳定性及滚动体验优化 (#790) 感谢 @yi124773651

• Fix(reader): 修复窗口关闭时的清理流程，正确停止服务器、释放 WebView2、关闭数据库并销毁窗口

• Fix(reader): 添加滚动防抖机制，优化翻页体验

• Fix(reader): 修复 Android 10+ 设备图片保存权限问题，移除不必要的存储权限请求 (#793)

• Fix(reader): 修复 Windows 端右键触发 WebView2 系统上下文菜单（返回、刷新、另存为、打印）的问题 (#746)

• Fix(reader): 修复图片内嵌大 max-width 导致宽度超出栏宽而显示到下一页的问题 (#742)

• Fix(reader): 修复禁用书籍自带样式时 epub 标题颜色不跟随主题色的问题，智能识别黑色标题并进行覆盖 (#749)

• Fix(toc): 目录中过长的章节标题现在会自动换行显示完整内容，不再以省略号截断 (#751)

• Fix(l10n): 修复首次启动时系统语言不受支持会错误回退到阿拉伯语的问题，改为回退到英文 (#785)

• Feat(reader): 页眉和页脚阅读信息支持分别设置左右边距与字体大小，并重构为共享的分区配置模型 (#791)

• Fix(l10n): 更新俄语翻译 app_ru.arb (#774) 感谢 @Xapitonov

• L10n: 完成罗马尼亚语翻译 (#714) 感谢 @Steinhagen

v1.14.0-beta.1

1.14.0

• Feat(reader): Add background image blur and opacity controls in reading settings (#753)

• Fix(reader): Fix background image effects not applying and switch reader background fill mode to cover (#753)

• Fix(reader): Fix RangeError crash when read theme color is null/invalid (#759)

• Feat(tts): Add Narrator option to text selection toolbar to start TTS from selected text (#794) Thanks @deskangel

• Feat(tts): Add click to pause/resume TTS playback on currently reading text (#794) Thanks @deskangel

• Feat(reader): Add Ctrl+[ and Ctrl+] page turning shortcuts on macOS, support Logitech Options+ mouse button mapping (#794) Thanks @deskangel

• Feat(ai): Add global AI RPM rate limiting in AI service layer

• Fix(ai): Fix unable to use AI for full-text translation

• Fix(tts): Fix TTS type selection having no effect when changed in reading interface (#794) Thanks @deskangel

• Fix(tts): Fix incorrect reading position after modifying pitch or rate (#794) Thanks @deskangel

• Fix(reader): Desktop resource lifecycle, WebView stability, and scroll UX optimization (#790) Thanks @yi124773651

• Fix(reader): Fix window close cleanup to properly stop server, dispose WebView2, close database, and destroy window

• Fix(ai): Fix AI crash prevention by adding guards for null webViewEnvironment on Windows

• Fix(reader): Add scroll debounce mechanism for smoother page turning experience

• Fix(l10n): Update Russian translation in app_ru.arb (#774) Thanks @Xapitonov

• Fix(l10n): Update full_description.txt (#775) Thanks @Xapitonov

• Docs: Add CLAUDE.md for AI assistant guidance (#792) Thanks @Anxcye

• Fix(toc): Show full chapter titles in TOC by wrapping long text instead of truncating with ellipsis (#751)

• L10n: Complete Romanian translation (#714) Thanks @Steinhagen

• Feat(reader): 阅读设置新增背景图模糊度和透明度调节 (#753)

• Fix(reader): 修复背景图效果不生效的问题，并将阅读器背景图填充方式调整为 cover (#753)

• Feat(tts): 在文本选择工具栏新增朗读选项，支持从选中文本开始 TTS (#794) 感谢 @deskangel

• Fix(reader): 修复阅读主题颜色为空/无效时的 RangeError 崩溃 (#759)

• Feat(tts): 新增点击正在朗读的文本以暂停/继续 TTS 播放 (#794) 感谢 @deskangel

• Feat(reader): macOS 新增 Ctrl+[ 和 Ctrl+] 翻页快捷键，支持 Logitech Options+ 鼠标按键映射 (#794) 感谢 @deskangel

• Feat(ai): 在 AI 服务层添加全局 AI 请求限流（RPM）

• Fix(ai): 修复无法使用 AI 进行全文翻译的问题

• Fix(tts): 修复在阅读界面切换 TTS 类型无效的问题 (#794) 感谢 @deskangel

• Fix(tts): 修复调整音调或语速后朗读位置不正确的问题 (#794) 感谢 @deskangel

• Fix(reader): 桌面端资源生命周期管理、WebView 稳定性及滚动体验优化 (#790) 感谢 @yi124773651

• Fix(reader): 修复窗口关闭时的清理流程，正确停止服务器、释放 WebView2、关闭数据库并销毁窗口

• Fix(ai): 修复 AI 崩溃问题，为 Windows 上空 webViewEnvironment 添加防护

• Fix(reader): 添加滚动防抖机制，优化翻页体验

• Fix(l10n): 更新俄语翻译 app_ru.arb (#774) 感谢 @Xapitonov

• Fix(l10n): 更新 full_description.txt (#775) 感谢 @Xapitonov

• Docs: 添加 CLAUDE.md 为 AI 助手提供指导 (#792) 感谢 @Anxcye

• Fix(toc): 目录中过长的章节标题现在会自动换行显示完整内容，不再以省略号截断 (#751)

• L10n: 完成罗马尼亚语翻译 (#714) 感谢 @Steinhagen

v1.12.0

1.12.0

• Feat(tts): Add OpenAI-compatible TTS service support (#653)

• Feat(tts): Add Aliyun TTS support with access key auth and auto endpoint

• Feat(tts): Add Aliyun voice list and descriptions

• Feat(reader): Add code syntax highlighting with 12 themes, 60+ languages

• Perf(reader): Optimize page turning smoothness and fix momentum scrolling issues on Android devices

• Fix(reader): Change default auto-translate selection setting to disabled

• Feat(reader): Add heading font size adjustment slider

• Feat(reader): Add customizable column threshold for auto single/double column switching

• Feat(reader): Add glassmorphism history navigation capsule in reader

• Feat(reader): Support manual control of history navigation visibility (no auto-hide)

• Feat(reader): Support manually selecting and remembering day/night background image preference

• Feat(reader): Support custom page turn area configuration with 3x3 grid

• Feat(reader): Add "Use Book Styles" toggle to preserve original book styling

• Feat(reader): Add toggle for mouse hover menu display at top/bottom edges

• Feat(reader): Add mouse wheel zoom support for image viewer

• Feat(bookshelf): Support displaying book title and author on default cover with responsive scaling

• Feat(ai): Support unlimited custom AI prompts with horizontal scrollable quick access

• Feat(ai): Support toggling AI panel position between right side and bottom in large screen mode

• Feat(sync): Add comprehensive WebDAV test (create, upload, download, delete) for connection validation

• Feat(window): Support window position and size memory on macOS

• Feat(import): Use original filename when book metadata title is missing or invalid

• Feat(import): Add detailed error message display for failed book imports with info dialog @navono

• UI(reader): Optimize reading history navigation button design

• Fix(reader): Fix image loading issue in non-standard EPUBs with missing manifest entries

• Fix(reader): Fix AI quick prompt chips overlapping issue

• Fix(reader): Fix PDF files with invalid TOC destinations failing to open

• Fix(reader): Fix reading info center alignment in header and footer

• Fix(reader): Fix search results scroll position not being preserved when returning from a result

• Fix(reader): Fix font-size adjustment not working for CSS keyword values (medium, large, etc.)

• Fix(statistics): Fix reading progress not being saved when opening books from continue reading tile

• Fix(sync): Fix WebDAV sync showing multiple direction selection dialogs

• Fix(ai_services): update Gemini service URL and model version

• Fix(window): Fix window becoming inaccessible after disconnecting secondary display

• Fix(sync): Fix WebDAV sync failure on Android devices with older SQLite versions

• Fix(import): Fix cross-platform path handling issues on Windows @navono

• Fix(bookshelf): Fix replace book file feature not converting TXT files to EPUB before replacement

• Fix(import): Enhance EPUB cover extraction with fallback strategies for non-standard formats

• Fix(bookshelf): Fix stacked view showing sharp corners when cover width is narrow (#727)

• Feat(reader): 添加代码语法高亮，支持 12 种主题, 60+ 编程语言

• Feat(tts): 增加 OpenAI 兼容 TTS 服务支持 (#653)

• Feat(tts): 增加阿里云 TTS 支持，使用 AccessKey 鉴权并支持自动就近 Endpoint

• Feat(tts): 增加阿里云音色列表与描述

• Perf(reader): 优化翻页流畅度，修复安卓设备惯性滚动问题

• Fix(reader): 修改自动翻译选中文本的默认设置为关闭

• Feat(reader): 添加标题字体大小调节滑块

• Feat(reader): 添加单双栏切换阈值自定义功能

• Feat(reader): 添加阅读器历史导航玻璃拟态胶囊

• Feat(reader): 支持手动控制历史导航显隐（不自动隐藏）

• Feat(reader): 支持手动选择并记忆日夜背景图偏好

• Feat(reader): 支持自定义翻页区域配置（3x3网格）

• Feat(reader): 添加"使用书籍样式"开关，保留书籍原始样式

• Feat(reader): 添加鼠标悬浮在顶部/底部时显示菜单的开关

• Feat(reader): 添加图片查看器鼠标滚轮缩放支持

• Feat(bookshelf): 支持在默认封面上显示书名和作者，支持响应式缩放

• Feat(ai): 支持无限制自定义AI提示词，快捷访问支持横向滚动

• Feat(ai): 支持切换 AI 面板位置（右侧/底部），适用于大屏模式

• Feat(sync): 添加 WebDAV 完整测试功能（创建、上传、下载、删除），全面验证连接

• Feat(window): 支持 macOS 记忆窗口位置和大小

• Fix(statistics): 修复从继续阅读卡片打开书籍时阅读进度无法保存的问题

• Feat(import): 当书籍元数据标题缺失或无效时，使用原始文件名

• Feat(import): 添加书籍导入失败时的详细错误信息显示和信息对话框 @navono

• UI(reader): 优化阅读历史导航按钮设计

• Fix(reader): 修复不规范 EPUB（manifest 缺失资源声明）中图片无法加载的问题

• Fix(reader): 修复AI快速提示重叠问题

• Fix(reader): 修复包含无效目录的 PDF 文件无法打开的问题

• Fix(reader): 修复页眉页脚阅读信息居中对齐问题

• Fix(reader): 修复点击搜索结果后重新打开搜索抽屉时滚动位置丢失的问题

• Fix(reader): 修复使用 CSS 关键字值（medium、large 等）的字体大小无法调节的问题

• Fix(sync): 修复 WebDAV 同步时出现多个方向选择弹窗的问题

• Fix(ai_services): 更新 Gemini 服务 URL 和模型版本

• Fix(window): 修复断开副屏后窗口位置不可访问的问题

• Fix(sync): 修复 Android 旧版 SQLite 设备 WebDAV 同步失败问题

• Fix(import): 修复 Windows 平台路径处理问题 @navono

• Fix(bookshelf): 修复替换书籍文件功能未将 TXT 文件转换为 EPUB 的问题

• Fix(import): 增强 EPUB 封面提取功能，支持非标准格式的多种回退策略

• Fix(bookshelf): 修复当封面宽度较低时书架重叠视图存在异常的直角边的问题 (#727)

v1.12.0-beta.3

1.12.0

• Feat(tts): Add OpenAI-compatible TTS service support (#653)

• Feat(tts): Add Aliyun TTS support with access key auth and auto endpoint

• Feat(tts): Add Aliyun voice list and descriptions

• Feat(reader): Add code syntax highlighting with 12 themes, 60+ languages

• Perf(reader): Optimize page turning smoothness and fix momentum scrolling issues on Android devices

• Fix(reader): Change default auto-translate selection setting to disabled

• Feat(reader): Add heading font size adjustment slider

• Feat(reader): Add customizable column threshold for auto single/double column switching

• Feat(reader): Add glassmorphism history navigation capsule in reader

• Feat(reader): Support manual control of history navigation visibility (no auto-hide)

• Feat(reader): Support manually selecting and remembering day/night background image preference

• Feat(reader): Support custom page turn area configuration with 3x3 grid

• Feat(reader): Add "Use Book Styles" toggle to preserve original book styling

• Feat(reader): Add toggle for mouse hover menu display at top/bottom edges

• Feat(reader): Add mouse wheel zoom support for image viewer

• Feat(bookshelf): Support displaying book title and author on default cover with responsive scaling

• Feat(ai): Support unlimited custom AI prompts with horizontal scrollable quick access

• Feat(ai): Support toggling AI panel position between right side and bottom in large screen mode

• Feat(sync): Add comprehensive WebDAV test (create, upload, download, delete) for connection validation

• Feat(window): Support window position and size memory on macOS

• Feat(import): Use original filename when book metadata title is missing or invalid

• Feat(import): Add detailed error message display for failed book imports with info dialog @navono

• UI(reader): Optimize reading history navigation button design

• Fix(reader): Fix image loading issue in non-standard EPUBs with missing manifest entries

• Fix(reader): Fix AI quick prompt chips overlapping issue

• Fix(reader): Fix PDF files with invalid TOC destinations failing to open

• Fix(reader): Fix reading info center alignment in header and footer

• Fix(reader): Fix search results scroll position not being preserved when returning from a result

• Fix(reader): Fix font-size adjustment not working for CSS keyword values (medium, large, etc.)

• Fix(statistics): Fix reading progress not being saved when opening books from continue reading tile

• Fix(sync): Fix WebDAV sync showing multiple direction selection dialogs

• Fix(ai_services): update Gemini service URL and model version

• Fix(window): Fix window becoming inaccessible after disconnecting secondary display

• Fix(sync): Fix WebDAV sync failure on Android devices with older SQLite versions

• Fix(import): Fix cross-platform path handling issues on Windows @navono

• Fix(bookshelf): Fix replace book file feature not converting TXT files to EPUB before replacement

• Fix(import): Enhance EPUB cover extraction with fallback strategies for non-standard formats

• Fix(bookshelf): Fix stacked view showing sharp corners when cover width is narrow (#727)

• Feat(reader): 添加代码语法高亮，支持 12 种主题, 60+ 编程语言

• Feat(tts): 增加 OpenAI 兼容 TTS 服务支持 (#653)

• Feat(tts): 增加阿里云 TTS 支持，使用 AccessKey 鉴权并支持自动就近 Endpoint

• Feat(tts): 增加阿里云音色列表与描述

• Perf(reader): 优化翻页流畅度，修复安卓设备惯性滚动问题

• Fix(reader): 修改自动翻译选中文本的默认设置为关闭

• Feat(reader): 添加标题字体大小调节滑块

• Feat(reader): 添加单双栏切换阈值自定义功能

• Feat(reader): 添加阅读器历史导航玻璃拟态胶囊

• Feat(reader): 支持手动控制历史导航显隐（不自动隐藏）

• Feat(reader): 支持手动选择并记忆日夜背景图偏好

• Feat(reader): 支持自定义翻页区域配置（3x3网格）

• Feat(reader): 添加"使用书籍样式"开关，保留书籍原始样式

• Feat(reader): 添加鼠标悬浮在顶部/底部时显示菜单的开关

• Feat(reader): 添加图片查看器鼠标滚轮缩放支持

• Feat(bookshelf): 支持在默认封面上显示书名和作者，支持响应式缩放

• Feat(ai): 支持无限制自定义AI提示词，快捷访问支持横向滚动

• Feat(ai): 支持切换 AI 面板位置（右侧/底部），适用于大屏模式

• Feat(sync): 添加 WebDAV 完整测试功能（创建、上传、下载、删除），全面验证连接

• Feat(window): 支持 macOS 记忆窗口位置和大小

• Fix(statistics): 修复从继续阅读卡片打开书籍时阅读进度无法保存的问题

• Feat(import): 当书籍元数据标题缺失或无效时，使用原始文件名

• Feat(import): 添加书籍导入失败时的详细错误信息显示和信息对话框 @navono

• UI(reader): 优化阅读历史导航按钮设计

• Fix(reader): 修复不规范 EPUB（manifest 缺失资源声明）中图片无法加载的问题

• Fix(reader): 修复AI快速提示重叠问题

• Fix(reader): 修复包含无效目录的 PDF 文件无法打开的问题

• Fix(reader): 修复页眉页脚阅读信息居中对齐问题

• Fix(reader): 修复点击搜索结果后重新打开搜索抽屉时滚动位置丢失的问题

• Fix(reader): 修复使用 CSS 关键字值（medium、large 等）的字体大小无法调节的问题

• Fix(sync): 修复 WebDAV 同步时出现多个方向选择弹窗的问题

• Fix(ai_services): 更新 Gemini 服务 URL 和模型版本

• Fix(window): 修复断开副屏后窗口位置不可访问的问题

• Fix(sync): 修复 Android 旧版 SQLite 设备 WebDAV 同步失败问题

• Fix(import): 修复 Windows 平台路径处理问题 @navono

• Fix(bookshelf): 修复替换书籍文件功能未将 TXT 文件转换为 EPUB 的问题

• Fix(import): 增强 EPUB 封面提取功能，支持非标准格式的多种回退策略

• Fix(bookshelf): 修复当封面宽度较低时书架重叠视图存在异常的直角边的问题 (#727)

v1.12.0-beta.2

1.12.0

• Feat(reader): Add code syntax highlighting with 12 themes, 60+ languages

• Perf(reader): Optimize page turning smoothness and fix momentum scrolling issues on Android devices

• Fix(reader): Change default auto-translate selection setting to disabled

• Feat(reader): Add heading font size adjustment slider

• Feat(reader): Add customizable column threshold for auto single/double column switching

• Feat(reader): Add glassmorphism history navigation capsule in reader

• Feat(reader): Support manual control of history navigation visibility (no auto-hide)

• Feat(reader): Support manually selecting and remembering day/night background image preference

• Feat(reader): Support custom page turn area configuration with 3x3 grid

• Feat(reader): Add "Use Book Styles" toggle to preserve original book styling

• Feat(reader): Add toggle for mouse hover menu display at top/bottom edges

• Feat(reader): Add mouse wheel zoom support for image viewer

• Feat(bookshelf): Support displaying book title and author on default cover with responsive scaling

• Feat(ai): Support unlimited custom AI prompts with horizontal scrollable quick access

• Feat(ai): Support toggling AI panel position between right side and bottom in large screen mode

• Feat(sync): Add comprehensive WebDAV test (create, upload, download, delete) for connection validation

• Feat(window): Support window position and size memory on macOS

• Feat(import): Use original filename when book metadata title is missing or invalid

• Feat(import): Add detailed error message display for failed book imports with info dialog @navono

• UI(reader): Optimize reading history navigation button design

• Fix(reader): Fix image loading issue in non-standard EPUBs with missing manifest entries

• Fix(reader): Fix AI quick prompt chips overlapping issue

• Fix(reader): Fix PDF files with invalid TOC destinations failing to open

• Fix(reader): Fix reading info center alignment in header and footer

• Fix(reader): Fix search results scroll position not being preserved when returning from a result

• Fix(reader): Fix font-size adjustment not working for CSS keyword values (medium, large, etc.)

• Fix(statistics): Fix reading progress not being saved when opening books from continue reading tile

• Fix(sync): Fix WebDAV sync showing multiple direction selection dialogs

• Fix(ai_services): update Gemini service URL and model version

• Fix(window): Fix window becoming inaccessible after disconnecting secondary display

• Fix(sync): Fix WebDAV sync failure on Android devices with older SQLite versions

• Fix(import): Fix cross-platform path handling issues on Windows @navono

• Fix(bookshelf): Fix replace book file feature not converting TXT files to EPUB before replacement

• Fix(import): Enhance EPUB cover extraction with fallback strategies for non-standard formats

• Fix(bookshelf): Fix stacked view showing sharp corners when cover width is narrow (#727)

• Feat(reader): 添加代码语法高亮，支持 12 种主题, 60+ 编程语言

• Perf(reader): 优化翻页流畅度，修复安卓设备惯性滚动问题

• Fix(reader): 修改自动翻译选中文本的默认设置为关闭

• Feat(reader): 添加标题字体大小调节滑块

• Feat(reader): 添加单双栏切换阈值自定义功能

• Feat(reader): 添加阅读器历史导航玻璃拟态胶囊

• Feat(reader): 支持手动控制历史导航显隐（不自动隐藏）

• Feat(reader): 支持手动选择并记忆日夜背景图偏好

• Feat(reader): 支持自定义翻页区域配置（3x3网格）

• Feat(reader): 添加"使用书籍样式"开关，保留书籍原始样式

• Feat(reader): 添加鼠标悬浮在顶部/底部时显示菜单的开关

• Feat(reader): 添加图片查看器鼠标滚轮缩放支持

• Feat(bookshelf): 支持在默认封面上显示书名和作者，支持响应式缩放

• Feat(ai): 支持无限制自定义AI提示词，快捷访问支持横向滚动

• Feat(ai): 支持切换 AI 面板位置（右侧/底部），适用于大屏模式

• Feat(sync): 添加 WebDAV 完整测试功能（创建、上传、下载、删除），全面验证连接

• Feat(window): 支持 macOS 记忆窗口位置和大小

• Fix(statistics): 修复从继续阅读卡片打开书籍时阅读进度无法保存的问题

• Feat(import): 当书籍元数据标题缺失或无效时，使用原始文件名

• Feat(import): 添加书籍导入失败时的详细错误信息显示和信息对话框 @navono

• UI(reader): 优化阅读历史导航按钮设计

• Fix(reader): 修复不规范 EPUB（manifest 缺失资源声明）中图片无法加载的问题

• Fix(reader): 修复AI快速提示重叠问题

• Fix(reader): 修复包含无效目录的 PDF 文件无法打开的问题

• Fix(reader): 修复页眉页脚阅读信息居中对齐问题

• Fix(reader): 修复点击搜索结果后重新打开搜索抽屉时滚动位置丢失的问题

• Fix(reader): 修复使用 CSS 关键字值（medium、large 等）的字体大小无法调节的问题

• Fix(sync): 修复 WebDAV 同步时出现多个方向选择弹窗的问题

• Fix(ai_services): 更新 Gemini 服务 URL 和模型版本

• Fix(window): 修复断开副屏后窗口位置不可访问的问题

• Fix(sync): 修复 Android 旧版 SQLite 设备 WebDAV 同步失败问题

• Fix(import): 修复 Windows 平台路径处理问题 @navono

• Fix(bookshelf): 修复替换书籍文件功能未将 TXT 文件转换为 EPUB 的问题

• Fix(import): 增强 EPUB 封面提取功能，支持非标准格式的多种回退策略

• Fix(bookshelf): 修复当封面宽度较低时书架重叠视图存在异常的直角边的问题 (#727)