Prepare for Hugging Face Spaces deployment

Major upgrade: 50-question benchmark, persistent analytics, paper-style TABLE II

Fix card rendering: safe_answer(), benchmark accuracy key, and badge blank-line bug

Major upgrade: auto-metrics, Plotly charts, Qwen2.5, PDF support, expanded KB

Improve UI: animated 3D background, spinning logo, and interactive quick questions

Add LLM vs RAG vs Fine-Tuning comparison project