feat(brain_answer): Qwen3-Reranker cross-encoder filter (opt-in)

Adds an opt-in cross-encoder rerank step between BM25 retrieval and LLM
synthesis. With BRAIN_RERANKER_URL set, brain_answer retrieves BM25
top-20, scores each excerpt against the query via Qwen3-Reranker on
Ollama, drops the "no" answers, and forwards up to 5 surviving sources
to the LLM. Unset, behaviour is unchanged (BM25 top-10 → LLM).

The reranker is a *filter*, not a re-ranker: Qwen3-Reranker emits a
binary yes/no token under its native chat template, and ties within the
"yes" set are broken by BM25 rank — what got retrieved first stays
ahead.

New package ingestion/internal/reranker:
- Client with URL, Model, HTTP fields.
- New(url, model) returns nil on empty url so callers can treat
  "feature disabled" as a single nil check.
- Score(ctx, query, docs) issues one /api/generate call per doc using
  the Qwen3-Reranker yes/no chat template (verbatim, because the model
  was trained on this exact wording). Parses the first non-think token.

Wiring:
- mcp.Server gains a WithReranker fluent setter to keep NewServer
  signature stable.
- brain_answer's BM25 limit jumps to 20 only when a reranker is wired,
  to give the filter something to do.
- cmd/server/main.go reads BRAIN_RERANKER_URL (+ optional
  BRAIN_RERANKER_MODEL, default dengcao/Qwen3-Reranker-0.6B:F16).

Tests cover: nil-on-empty-url, ordered yes/no scoring, request shape
(model, prompt contents, yes/no template), ambiguous response → 0,
empty doc slice, upstream-error propagation, plus an end-to-end
brain_answer integration that proves only the relevant note reaches the
LLM when noise.md is rejected.

Closes hyperguild#7.

ingestion/internal/mcp/tools_answer_test.go

@@ -3,14 +3,17 @@ package mcp_test
 import (
 	"context"
 	"encoding/json"
+	"io"
 	"net/http"
 	"net/http/httptest"
 	"os"
 	"path/filepath"
+	"strings"
 	"testing"
 
 	"github.com/mathiasbq/hyperguild/ingestion/internal/mcp"
 	"github.com/mathiasbq/hyperguild/ingestion/internal/pipeline"
+	"github.com/mathiasbq/hyperguild/ingestion/internal/reranker"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -46,6 +49,55 @@ func callTool(t *testing.T, ts *httptest.Server, name string, arguments map[stri
 	return out
 }
 
+func TestBrainAnswer_RerankerFiltersBeforeLLM(t *testing.T) {
+	brainDir := t.TempDir()
+	wikiDir := filepath.Join(brainDir, "wiki")
+	require.NoError(t, os.MkdirAll(wikiDir, 0o755))
+	// Two notes — both BM25-match the query, but only one is truly relevant.
+	require.NoError(t, os.WriteFile(filepath.Join(wikiDir, "good.md"), []byte(
+		"---\ntitle: Pass-rate Logging\n---\nPass-rate logging tracks skill invocations.",
+	), 0o644))
+	require.NoError(t, os.WriteFile(filepath.Join(wikiDir, "noise.md"), []byte(
+		"---\ntitle: Pass-rate Tangent\n---\nPass-rate appears here too but as a tangent.",
+	), 0o644))
+
+	// Fake Ollama reranker: yes only when prompt contains "tracks skill invocations".
+	rrSrv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		raw, _ := io.ReadAll(r.Body)
+		yes := strings.Contains(string(raw), "tracks skill invocations")
+		ans := "no"
+		if yes {
+			ans = "yes"
+		}
+		_ = json.NewEncoder(w).Encode(map[string]any{"response": ans, "done": true})
+	}))
+	defer rrSrv.Close()
+
+	// LLM mock captures the rendered sources so we can assert what reached it.
+	var sawSources string
+	llm := func(_ context.Context, _, user string) (string, error) {
+		sawSources = user
+		return "answer text", nil
+	}
+
+	srv := mcp.NewServer(brainDir, nil, nil, llm).
+		WithReranker(reranker.New(rrSrv.URL, "qwen3"))
+	ts := httptest.NewServer(srv)
+	defer ts.Close()
+
+	rpc := callTool(t, ts, "brain_answer", map[string]any{"query": "pass-rate logging"})
+	require.Nil(t, rpc["error"])
+
+	content := rpc["result"].(map[string]any)["content"].([]any)[0].(map[string]any)["text"].(string)
+	var result map[string]any
+	require.NoError(t, json.Unmarshal([]byte(content), &result))
+	sources := result["sources"].([]any)
+	require.Len(t, sources, 1, "reranker should drop noise.md")
+	assert.Equal(t, "wiki/good.md", sources[0])
+	assert.Contains(t, sawSources, "good.md")
+	assert.NotContains(t, sawSources, "noise.md")
+}
+
 func TestBrainAnswer_NoLLM(t *testing.T) {
 	srv := mcp.NewServer(t.TempDir(), nil, nil, nil)
 	ts := httptest.NewServer(srv)