fix: contextual retrieval latency — max_chunks gate + prompt trim + model update

eval: v1.3.0 clean run (25 samples) — recall=0.627, relevancy=0.837; fix eval script max_tokens+print

fix: eval script — max_tokens 500→1024, max_workers 2→1, timeout 120→180 to reduce RAGAS failures

feat: add --contextual flag to eval script for contextual retrieval eval run

feat: Stage 8 — eval dashboard, mandatory web search, docs update

fix: update index.html title and benchmark script to Prism branding

feat: rebrand FinRAG → Prism (collection names, system prompt, UI branding)

chore: run full RAGAS eval — all 4 metrics now populated

chore: update RAGAS benchmark with real scores (faithfulness=1.0, answer_relevancy=0.90)

feat: RAGAS benchmark driven by JSON, local runner script

Add multi-K precision evaluation support

Add Dockerfile and Render config