feat(brain): hybrid BM25 + pgvector retrieval (opt-in)

Wires nomic-embed-text (iguana ollama) + pgvector on the shared
postgres18 into brain_query / brain_answer via Reciprocal Rank Fusion.
Pure BM25 stays the default; setting BRAIN_PG_DSN and BRAIN_EMBED_URL
together opts in. Setting one without the other is misconfiguration →
exit 1.

New packages:

- internal/embed
  Client.Embed(ctx, text) → []float32 via POST {URL}/api/embed.
  Defaults to nomic-embed-text:latest (768 dim). nil-on-empty-URL so
  callers gate on a single nil check.

- internal/vectorstore
  PGStore wraps a pgxpool against postgres18. Init creates
  brain_embeddings(path PK, vector(768), updated_at) + HNSW cosine
  index idempotently. Upsert / Delete / Search / KnownPaths.
  Sync(brainDir, store, embedder) diffs brain/wiki/ against the store
  and upserts new files / deletes removed ones; StartSync runs it on
  a ticker (default 300s). Integration tests gated by BRAIN_PG_TEST_DSN.

- scripts/brain-embeddings-init.sql
  One-time DBA setup: brain DB, brain_app role, vector extension,
  GRANTs. Idempotent.

Search layer:

- search.QueryOptions gains Vector + Embedder fields.
- QueryContext is the cancellable variant; Query stays for callers.
- When both are set, BM25 (top-N) and pgvector (top-4N) candidates
  merge via Reciprocal Rank Fusion (k=60, Cormack et al. 2009 — no
  tuning knob, robust to scale differences between rankers).
- Vector-only hits are hydrated from disk so callers see uniform
  Result records (path, title, excerpt, wing, hall, score).
- Wing/hall filters still apply to vector candidates via path-prefix.
- On embedder/vector errors the search falls back to BM25 — embedding
  outage degrades quality but doesn't take the brain offline.

MCP wiring:

- mcp.Server.WithHybridRetrieval(v, e) opt-in setter, same shape as
  WithReranker.
- brainQuery and brainAnswer pass the wired vector/embedder through
  to search.QueryContext.

REST:

- POST /backfill-embeddings drives Sync synchronously. Returns
  {added, deleted, errors[]}. 503 when feature is unconfigured.

cmd/server/main.go:

- BRAIN_PG_DSN + BRAIN_EMBED_URL together enable hybrid; one alone
  → exit 1.
- vectorAdapter bridges *PGStore (returns []Hit) to
  search.VectorSearcher (which takes []VectorHit) without either
  package importing the other.
- BRAIN_EMBED_SYNC_INTERVAL (default 300s) controls the background
  Sync ticker.

Backend pivot from Qdrant to pgvector recorded in DECISIONS.md
2026-05-18 (supersedes 2026-04-08): postgres18 already runs in
databases/ ns, Qdrant was never deployed, one engine beats two.

Dependency: github.com/jackc/pgx/v5 — modern, native pgvector via
parametric vector literals.

Tests:
- embed.Client: empty-URL nil, request shape, dimension, upstream
  error propagation, empty-text rejection.
- vectorstore.PGStore: dimension validation (unit); upsert/search/
  KnownPaths (integration, BRAIN_PG_TEST_DSN-gated).
- vectorstore.Sync: adds new files, skips known, deletes
  disappeared, skips _index.md, no-op when nil, collects embedder
  errors.
- search.Query: hybrid promotes vector-only hits via RRF; falls
  back to BM25 on embedder error.

Closes hyperguild#8.

ingestion/internal/mcp/handlers.go

@@ -144,11 +144,13 @@ func (s *Server) brainQuery(ctx context.Context, args json.RawMessage) (json.Raw
 	if a.Limit == 0 {
 		a.Limit = 5
 	}
-	results, err := search.Query(s.brainDir, search.QueryOptions{
-		Query: a.Query,
-		Limit: a.Limit,
-		Wing:  a.Wing,
-		Hall:  a.Hall,
+	results, err := search.QueryContext(ctx, s.brainDir, search.QueryOptions{
+		Query:    a.Query,
+		Limit:    a.Limit,
+		Wing:     a.Wing,
+		Hall:     a.Hall,
+		Vector:   s.vector,
+		Embedder: s.embedder,
 	})
 	if err != nil {
 		return nil, fmt.Errorf("search: %w", err)

ingestion/internal/mcp/server.go

@@ -11,6 +11,7 @@ import (
 
 	"github.com/mathiasbq/hyperguild/ingestion/internal/pipeline"
 	"github.com/mathiasbq/hyperguild/ingestion/internal/reranker"
+	"github.com/mathiasbq/hyperguild/ingestion/internal/search"
 )
 
 type request struct {
@@ -39,6 +40,8 @@ type Server struct {
 	llm       pipeline.CompleteFunc
 	answerLLM pipeline.CompleteFunc // nil = brain_answer and brain_classify unavailable
 	reranker  *reranker.Client      // nil = no rerank, BM25 top-10 → LLM
+	vector    search.VectorSearcher // nil = BM25-only retrieval
+	embedder  search.Embedder       // nil = BM25-only retrieval
 }
 
 // NewServer constructs a Server bound to brainDir. pipelineCfg supplies the
@@ -61,6 +64,15 @@ func (s *Server) WithReranker(r *reranker.Client) *Server {
 	return s
 }
 
+// WithHybridRetrieval wires the embedding store and embedder so
+// brain_query and brain_answer run BM25 + pgvector merged via RRF
+// instead of BM25 alone. Either nil disables hybrid mode.
+func (s *Server) WithHybridRetrieval(v search.VectorSearcher, e search.Embedder) *Server {
+	s.vector = v
+	s.embedder = e
+	return s
+}
+
 func (s *Server) ServeHTTP(w http.ResponseWriter, r *http.Request) {
 	// MCP streamable HTTP: GET establishes the SSE stream for server-to-client events.
 	if r.Method == http.MethodGet {

ingestion/internal/mcp/tools_answer.go

@@ -67,7 +67,12 @@ func (s *Server) brainAnswer(ctx context.Context, args json.RawMessage) (json.Ra
 	if s.reranker != nil {
 		bm25Limit = 20
 	}
-	results, err := search.Query(s.brainDir, search.QueryOptions{Query: a.Query, Limit: bm25Limit})
+	results, err := search.QueryContext(ctx, s.brainDir, search.QueryOptions{
+		Query:    a.Query,
+		Limit:    bm25Limit,
+		Vector:   s.vector,
+		Embedder: s.embedder,
+	})
 	if err != nil {
 		return nil, fmt.Errorf("search: %w", err)
 	}