OrigRead Desktop v1.1.0

新增功能

• 新增全局搜索，使用 Ctrl / Cmd + Shift + F 搜索本机已缓存文章的标题、摘要和正文；搜索结果显示上下文，点击即可跳转阅读。
• 新增 AI 网站规则和 JSON 规则生成流程，可选择已配置的 AI 服务和模型，并在保存前查看真实解析结果。
• 网站列表规则和正文规则分阶段生成、验证；正文规则失败不会阻塞列表规则保存，并提供通用解析和原网页兜底。
• 来源解析方式按自动选择、已配置规则和内置智能识别分组显示，规则启用状态与来源设置保持同步。
• 用户手册已内置到桌面端的提示与支持页面。

问题修复

• 修复来源规则与实际启用状态显示不同步的问题。
• 改进网站和 JSON 来源的正文提取、HTML 清理和解析失败处理。
• 改进更新检查对 Release 标签和页面地址的校验，并支持代理失败后回退官方地址。

体验优化

• AI 规则生成过程显示当前阶段、模型、服务、验证结果和失败原因。
• 优化 AI 服务和模型选择界面，模型搜索框不再自动填入当前模型。
• 优化网站解析方式页面的信息分组和规则状态展示。

---

New Features

• Added global search with Ctrl / Cmd + Shift + F for cached article titles, summaries, and full text, including context snippets and click-to-open results.
• Added AI-assisted Website Rule and JSON Rule generation with configurable provider and model selection and real parsing previews before saving.
• Website list rules and content rules are generated and validated in separate stages; a content-rule failure no longer blocks saving the list rule and falls back to general parsing or the original page.
• Source parsing choices are grouped into automatic selection, configured rules, and built-in detection, with enabled states synchronized with source settings.
• Embedded the user guide in the desktop Help and Support section.

Bug Fixes

• Fixed mismatches between configured rule enabled states and the source parsing UI.
• Improved content extraction, HTML cleanup, and failure handling for Website and JSON sources.
• Improved update checks by validating Release tags and page URLs and falling back to the official endpoint when the proxy fails.

Experience Improvements

• AI rule generation now shows its current stage, provider, model, validation result, and user-facing failure reason.
• Improved AI provider and model selection so the model search field is not prefilled with the current model.
• Improved grouping and status presentation in the website parsing method screen.

Full Changelog: v1.0.1...v1.1.0