refactor: extract shared query parser module, clean up imports

fix: reranker degradation, skip_reranker bug, eval latency, branch cleanup

feat: full 100k index pipeline - switch from samples to full offset index

feat(eval): add ground truth generation, real evaluation metrics, remove demo mode

feat(scripts): rewrite build_indexes and evaluate with streaming and demo mode

feat: Phase 13 evaluation — P@k, R@k, MRR, nDCG, latency stats

chore: create directory structure with __init__.py files and placeholder scripts