Dcup V-1.2.0

[aliamerj]

• new RAG engine

Query 
 ↓ Expand queries with LLM (better recall)
 ↓ Parallel dense + sparse embedding (hybrid search)
 ↓ Hybrid retrieval (best of both worlds)
 ↓ Deduplicate (remove redundant results)
 ↓ Normalize (score consistency)
 ↓ Cross-encoder rerank (precision boost)
 ↓ Context reconstruction (better answers)
 ↓ Context merge (coherent output)
 ↓ LLM answer (final response)

• clean up and better UI
• chat
• PWA
• connection notion