lint

fix: preserve gemini thought_signatures required for tool calls

feat: enhance LLM formatting capabilities and improve logging in various components

review: ship all 15 quick-win fixes from REVIEW.md via TDD

feat: add youtube_frame_at tool, improve vision URL validation, silence observability logs, and update caveman mode configuration.

feat: implement GAIA benchmark evaluation runner with strict output formatting and vision tool integration.

feat: implement tool error cooldown, caveman system prompt mode, and iteration fail-safe logic

feat: implement web search and file processing tools with robust retrieval fallbacks