Composes the existing LexicalRetriever (P2-2) with a new VectorRetriever
wrapper around LanceVectorStore (P3-3) into a single Retriever that
dispatches by SearchMode. For SearchMode::Hybrid, fuses lexical and
vector candidates via Reciprocal Rank Fusion and populates the full
RetrievalDetail per SearchHit so kb search --explain can attribute
scores back to each side.

Public surface (kb-search crate):
- pub struct VectorRetriever — Arc<dyn VectorStore + Send + Sync>,
  Arc<dyn Embedder>, Arc<SqliteStore>, IndexVersion at construction.
- pub struct HybridRetriever { lexical, vector, fusion, k }.
- pub enum FusionPolicy { Rrf { k_rrf: u32 } }.

VectorRetriever:
- Embeds query.text as EmbeddingKind::Query before delegating to
  VectorStore::search(query_vec, query.k * 2, &query.filters). Over-
  fetches by ×2 for filter losses; LanceVectorStore applies the
  filters internally so they propagate naturally.
- Hydrates each VectorHit into a full SearchHit by joining on
  chunk_id in a single IN-clause batch (no N+1): doc_path,
  section_label, chunker_version, source_spans for citation, plus
  embedding_model from embedder.model_id().
- Snippet trimmed to config.search.snippet_chars (vector mode lacks
  FTS5 highlighting; chunk text prefix is the next-best signal).
- Citation built from the chunk's first source span via the shared
  citation_helper module — extracted from lexical.rs so both
  retrievers compute citations identically (Byte/empty fallback to
  Line{1,1} preserved with tracing::warn).
- RetrievalDetail.method = Vector for standalone calls; both
  fusion_score and vector_score set to the LanceVectorStore-shifted
  cosine score; lexical_* None.

HybridRetriever:
- Lexical / Vector modes delegate 1:1 — no rebuild of RetrievalDetail.
- Hybrid mode runs both retrievers with k * 2 fanout, fuses with
  RRF (score(c) = Σ 1/(k_rrf + rank_m(c))), sorts fused-score DESC
  with deterministic tiebreaker (lex_rank ASC then chunk_id ASC),
  takes top query.k. Fusion math runs in f64 throughout; cast to
  f32 only at the SearchHit boundary where bounded magnitude (≤
  ~0.033 at k_rrf=60) makes f32 precision sufficient for ranking.
- Per-hit lexical preferred for snippet/citation/heading_path/
  chunker_version/embedding_model when the chunk appears in both
  retrievers — FTS5 highlighting is more user-relevant than vector's
  truncated text. Vector-only chunks fall through to vector hit data.
- index_version returns format!("hybrid:{}+{}", lex_iv, vec_iv) at
  construction; mismatched lex/vec versions trigger a tracing::warn
  so users notice stale indexes (spec line 143).

kb-search additions:
- citation_helper.rs — pub(crate) citation_from_first_span shared
  between lexical and vector retrievers. Extracted from lexical.rs;
  no behavior drift.

Tests (38 default + 3 ignored):
- 12 unit tests in hybrid.rs covering RRF math (1/61 + 1/62 within
  f32 epsilon × 10 tolerance), lexical/vector mode delegation, hybrid
  preserves single-side hits with the missing side's RetrievalDetail
  None, deterministic tiebreaker on identical fused scores, composite
  index_version, mismatched-version warn at construction.
- 2 unit tests in vector.rs covering the snippet-prefix and citation
  fallback paths.
- 11 unit tests in lexical.rs (unchanged from P2-2).
- 13 lexical integration tests (unchanged).
- 3 #[ignore] AVX-gated hybrid integration tests: disjoint-corpus
  recall (lex returns A,B; vec returns C,D; hybrid returns all 4),
  determinism over two queries, snapshot stability against
  tests/fixtures/search/hybrid/run-1.json. Snapshot fixture was
  regenerated against this branch on an AVX-enabled VM and contains
  4 real chunks (c1/c2 lex+vec, c3/c4 vec-only).
- KB_UPDATE_SNAPSHOTS=1 path now panics after writing instead of
  silently passing — matches the P3-2/P3-3 fail-loud-instead-of-
  silent-pass philosophy.

Allowed deps respected (kb-core, kb-config, kb-store-sqlite,
kb-store-vector, kb-embed, tracing, thiserror) plus pre-existing
kb-search deps from P2-2 (rusqlite, globset, serde_json, anyhow).
kb-embed-local does NOT appear — VectorRetriever takes Arc<dyn Embedder>
trait object; the concrete adapter is runtime-injected by kb-app.

Out of scope: reranker (P+), score calibration across modes (RRF is
rank-comparable so absolute calibration is P+), multimodal retrieval
(P6+).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

@@ -0,0 +1,74 @@
+//! Shared helpers for building `kb_core::Citation` values from a

@@ -0,0 +57,4 @@
+/// For chunks that appear in both retrievers, the lexical-side hit
+/// supplies `snippet`, `citation`, `heading_path`, `chunker_version`,
+/// and `embedding_model` — lexical search has FTS5 highlighting that's
+/// more user-relevant than the vector retriever's truncated text.

@@ -0,0 +87,4 @@
+        };
+        // Surface mismatched index_version up front so users see it
+        // (e.g. lexical at v2, vector at v1 means a stale index that
+        // the user should refresh). Spec line 144 calls this out as

@@ -0,0 +197,4 @@
+        struct Scored {
+            chunk_id: String,
+            rrf: f64,
+            lex_rank: Option<u32>,

@@ -0,0 +227,4 @@
+        // total_cmp keeps the sort stable under future tweaks).
+        scored.sort_by(|a, b| {
+            b.rrf
+                .total_cmp(&a.rrf)

@@ -0,0 +248,4 @@
+                (Some((_, lex)), _) => lex.clone(),
+                (None, Some((_, vec))) => vec.clone(),
+                // `all_ids` is the union of `lex_index` and
+                // `vec_index` keys, so this arm cannot fire.

@@ -0,0 +217,4 @@
+        .filter(|id| seen.insert(*id))
+        .collect();
+
+    let placeholders = vec!["?"; unique.len()].join(",");

@@ -0,0 +1,42 @@
+[

@@ -0,0 +167,4 @@
+        // philosophy as P3-2's `SNAPSHOT_HASH_BASELINE = 0` and P3-3's
+        // placeholder fixture guards.
+        panic!(
+            "[snapshot] regenerated {}, re-run without KB_UPDATE_SNAPSHOTS to verify pin",