UX overhaul + Gradio migration + remove llama.cpp

cache bust to force clean HF rebuild

Remove llama-cpp-python and lighten Dockerfile; Modal handles inference

fix: restore llama.cpp with source build, use fine-tuned GGUF model

refactor: remove llama-cpp entirely, use deterministic mock-only mode

fix: force llama-cpp-python source build to fix musl/glibc mismatch on HF Spaces

fix: cache bust Docker layer to force libgomp1/libgfortran5 install

fix: install libgomp1 + libgfortran5 so the prebuilt llama-cpp wheel can load

fix: harden LLM calls, cast numpy to native floats, pin llama-cpp-python wheel

Space: prebuilt llama-cpp-python wheel, minimal Dockerfile

Add model download, split requirements, prepare for Modal training

Phase 2/3: Gradio Server backend, CRT frontend, engine, agents, mentor, tests, CI/CD