fix: correct author name formatting in multiple files

Enhance YouTube video processing with transcript and frame analysis modes

refactor: rename runtime folders with underscore prefix

feat: consolidate LLM logs to single session file

feat: add Chain of Thought for LLM synthesis debugging

feat: standardize logging with console + file separation

refactor: compress console logs for clarity

feat: export LLM synthesis context to cache for debugging

debug: log full transcript content for debugging

feat: add transcript caching and upgrade synthesis model

feat: system error field, calculator fix, target task IDs, course vs GAIA docs

Update Groq model to openai/gpt-oss-120b

Fix: Read LLM provider config at runtime for UI updates

Add UI-based LLM provider selection for cloud testing

Groq Model Update

Stage 5: Performance optimization - retry logic, Groq integration, improved prompts

Feat: Add HuggingFace Inference API as free LLM fallback tier

Fix: Make LLM provider configurable (Gemini primary, Claude fallback)

Stage 3: Core Logic Implementation - LLM Integration