feat(rag): optimize RAG pipeline — JSON-Mode, CoT, Hybrid Search, Re-Ranking, Cross-Reg Dedup, chunk 1024

Phase 1 (LLM Quality):
- Add format=json to all Ollama payloads (obligation_extractor, control_generator, citation_backfill)
- Add Chain-of-Thought analysis steps to Pass 0a/0b system prompts

Phase 2 (Retrieval Quality):
- Hybrid search via Qdrant Query API with RRF fusion + automatic text index (legal_rag.go)
- Fallback to dense-only search if Query API unavailable
- Cross-encoder re-ranking with BGE Reranker v2 (RERANK_ENABLED=false by default)
- CPU-only PyTorch dependency to keep Docker image small

Phase 3 (Data Layer):
- Cross-regulation dedup pass (threshold 0.95) links controls across regulations
- DedupResult.link_type field distinguishes dedup_merge vs cross_regulation
- Chunk size defaults updated 512/50 → 1024/128 for new ingestions only
- Existing collections and controls are NOT affected

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

admin-compliance/__tests__/ingest-industry-compliance.test.ts

@@ -48,12 +48,12 @@ describe('Ingestion Script: ingest-industry-compliance.sh', () => {
       expect(scriptContent).toContain('chunk_strategy=recursive')
     })
 
-    it('should use chunk_size=512', () => {
-      expect(scriptContent).toContain('chunk_size=512')
+    it('should use chunk_size=1024', () => {
+      expect(scriptContent).toContain('chunk_size=1024')
     })
 
-    it('should use chunk_overlap=50', () => {
-      expect(scriptContent).toContain('chunk_overlap=50')
+    it('should use chunk_overlap=128', () => {
+      expect(scriptContent).toContain('chunk_overlap=128')
     })
 
     it('should validate minimum file size', () => {