Add provenance fields to /health response

Document measured API p95 latency in README

Remove unused shap dependency

Add .dockerignore to keep build context clean

Persist rate limiter state across crash-restarts, harden state loading

Auto-restart FastAPI backend under supervisord if it crashes

Condense multi-line notebook comments; re-run to refresh output

Disclose recommendation-diversity finding and AMD calibration-set shortfall

Serve FastAPI and Streamlit together; UI calls API over HTTP

Add infeasibility response shape (top_recommendation/infeasibility)

Add API request-id tracing, rate limiting, and response provenance

More cleanup

Code cleanup

Close power-efficiency Pareto gap, add training-data disclosure, harden data lineage, cut Docker image 84%

Add MIT license, refresh README accuracy numbers post-calibration

GPU Perf Prophet v1 — cross-vendor GPU performance forecasting & recommendation engine