v0.6.0

What's Changed

• feat: LM Studio backend (local OpenAI-compatible, privacy-first) by @marlian in #41
• docs: add LM Studio to README, GitHub topics, privacy-first copy by @marlian in #42
• feat: LM Studio model listing and selection via /models + /model by @marlian in #43
• feat: migrate LM Studio to native v1 API (privacy-first, stateful, MCP-ready) by @marlian in #44
• docs: LM Studio native API architecture + Privacy section in README by @marlian in #45
• fix: truncate LM Studio model slugs for Telegram 64-byte callback_data limit by @marlian in #46
• fix: harden LM Studio streaming and model callbacks by @marlian in #47
• docs: align docs and changelog with LM Studio backend (PRs #41–#47) by @marlian in #48
• chore: release v0.5.0 by @marlian in #49
• chore: update [Unreleased] comparison link to v0.5.0 by @marlian in #50
• chore(deps): bump dotenv from 17.4.1 to 17.4.2 by @dependabot[bot] in #53
• chore(deps): bump softprops/action-gh-release from 2 to 3 by @dependabot[bot] in #51
• chore(deps): bump better-sqlite3 from 12.8.0 to 12.9.0 by @dependabot[bot] in #52
• feat: reload agent registry on /agents and /agent by @marlian in #54
• feat: add /restart command for supervisor-driven restart by @marlian in #55
• chore: release v0.6.0 by @marlian in #56

Full Changelog: v0.4.1...v0.6.0