feat: Add multi-type embedding backends for cascade retrieval

- Implemented BinaryEmbeddingBackend for fast coarse filtering using 256-dimensional binary vectors.
- Developed DenseEmbeddingBackend for high-precision dense vectors (2048 dimensions) for reranking.
- Created CascadeEmbeddingBackend to combine binary and dense embeddings for two-stage retrieval.
- Introduced utility functions for embedding conversion and distance computation.

chore: Migration 010 - Add multi-vector storage support

- Added 'chunks' table to support multi-vector embeddings for cascade retrieval.
- Included new columns: embedding_binary (256-dim) and embedding_dense (2048-dim) for efficient storage.
- Implemented upgrade and downgrade functions to manage schema changes and data migration.

codex-lens/src/codexlens/search/hybrid_search.py

@@ -40,11 +40,20 @@ from codexlens.search.ranking import (
     get_rrf_weights,
     reciprocal_rank_fusion,
     rerank_results,
+    simple_weighted_fusion,
     tag_search_source,
 )
 from codexlens.storage.dir_index import DirIndexStore
 
 
+# Three-way fusion weights (FTS + Vector + SPLADE)
+THREE_WAY_WEIGHTS = {
+    "exact": 0.2,
+    "splade": 0.3,
+    "vector": 0.5,
+}
+
+
 class HybridSearchEngine:
     """Hybrid search engine with parallel execution and RRF fusion.
 
@@ -193,9 +202,22 @@ class HybridSearchEngine:
             if source in results_map
         }
 
-        with timer("rrf_fusion", self.logger):
+        # Determine fusion method from config (default: rrf)
+        fusion_method = "rrf"
+        rrf_k = 60
+        if self._config is not None:
+            fusion_method = getattr(self._config, "fusion_method", "rrf") or "rrf"
+            rrf_k = getattr(self._config, "rrf_k", 60) or 60
+
+        with timer("fusion", self.logger):
             adaptive_weights = get_rrf_weights(query, active_weights)
-            fused_results = reciprocal_rank_fusion(results_map, adaptive_weights)
+            if fusion_method == "simple":
+                fused_results = simple_weighted_fusion(results_map, adaptive_weights)
+            else:
+                # Default to RRF
+                fused_results = reciprocal_rank_fusion(
+                    results_map, adaptive_weights, k=rrf_k
+                )
 
         # Optional: boost results that include explicit symbol matches
         boost_factor = (