Replace HHEM with sentence-level NLI, add claim decomposition and drift detection

Fix compat, bugs, and types; expand retail KB

Faithfulness: mean sentence scoring, strip chunk title prefix, lower threshold to 0.35

Address Gate 5 audit gaps

Add Makefile, HTML eval report generator, gitignore for reports

Add L2 batch evaluator and architecture documentation

Add full RAG evaluation pipeline with L1 metrics and UI