feat: overhaul model routing — GLM-5.1 primary, strict 4-model fallback chain, purge all stale refs

fix: pass HF_API_KEY as HF_TOKEN to avoid Hub rate limits when loading models

feat: implement ChatGPT style UI, streaming generation, math/code blocks, citations & multi-model fallback

feat: ResearchPilot v2 — Next.js frontend, 358k embeddings pipeline, LaTeX math rendering, dual-GPU Kaggle scaling

feat: RAG pipeline complete - ResearchPilot end-to-end working

feat: data ingestion and processing pipeline complete