fix: trust summary resilience, Unicode in reason strings, full GitHub README

tune STALE threshold to 0.50 — better Cat B recall (44% vs 2% at 0.40)

Phase 2.6: add per-paper trust summary to synthesizer output — addresses Armin's explainability request

Phase 2.5: wire reliability scorer into critic — edge reliability replaces mean-age STALE threshold

Phase 2.2-2.3: add doi to Paper, populate from S2 externalIds, integrate OpenAlex into retriever

Phase 1.2: fix STALE-before-CONTRADICTED bug — both checks now run, CONTRADICTED wins when both fire

Phase 13: HF Spaces deploy ready - verdict logging, clean requirements

Phase 9 polish: structured output format, HTML UI components, citation highlighting

Phase 7: synthesizer agent, claim extraction, session write, MD export

Phase 6: critic agent, 4-verdict taxonomy, contradiction detection, question rewriter

Phase 5: retriever agent, S2 keyword conversion, hybrid scoring, citation graph

Phase 4: planner agent with temporal decomposition and session context

Phase 2: state schema, SQLite session memory, data dir