v0.1.0

What's Changed

• feat: PDF/DOCX/ODT extraction, multimodal vision chat, STT/TTS by @SimonWaldherr with @Copilot in #3
• Add minimal extensible connector system (HTTP + SQL, tool registry, MCP) by @SimonWaldherr with @Copilot in #4
• Add multimodal vision support, more model providers, fix API key auth, admin/security features, and improved tool usage by @SimonWaldherr with @Copilot in #5
• Upgrade tinyRAG to streaming-first inline-tool architecture with XML protocol by @SimonWaldherr with @Copilot in #6
• Implement tinyRAG R³: governed retrieval, provenance-first citations, policy-driven tool persistence, and security hardening by @SimonWaldherr with @Copilot in #7
• feat: minimalist Web-UI refresh + universal hover tooltips by @SimonWaldherr with @Copilot in #8
• Standardize backends: vectorChunkStore abstraction, llmster support, multilingual embed hints by @SimonWaldherr with @Copilot in #9

Full Changelog: v0.0.3...v0.1.0