update documentations and readme

Change USE_REAL_AI to deferred validation via property

Fix DLQ retry: legacy stage handling and publish bookkeeping

Fix Round 3 code review findings: DLQ stages, publish retry, USE_REAL_AI

Fix 2nd round review findings: DLQ publish, stage names, docs

Fix code review findings: DLQ, AI mode, status mapping, caching

Fix code quality issues from project scan

Fix queue tests to match actual implementation

Add tests, update docs, and add code comments

Complete Phase 5 MVP: Load testing, resource limits, Docker optimization

Complete Phase 2 & 5: Query endpoints, rate limiting, DLQ management

Implement Phase 4: Real AI model evaluation

Sync documentation with actual implementation status

Implement Phase 3: Real Hyperliquid enrichment with signal validation

Implement phase 1