measure what each claimant on an utterance reports (V-565)

Two reporting tests over the 91-case RU fixture, no ratchet: a ratchet here
would freeze a number nobody has decided to hold.

TestStage0Contention runs the 21 grammars one at a time instead of stopping
at the first match. One case of 91 draws two, ru-query-019, where
calendar-query beats agenda-query by list position alone.

TestONNXClaimConfidenceDistribution buckets the reported confidence by the
layer that produced it. Stage 0 is 20/20 at a hardcoded 1.0. The classifier
scores 62% below its median and 62% above, across a cosine range of 0.859
to 0.942, with a top-two margin of p50 0.009. The float is not a confidence.

newBaselineClassifier and baselineGrammars split out of newBaselineRouter so
the measurement runs the same rules the daemon runs. TestONNXBaseline is
unchanged at 64/91.

internal/router/eval/eval_test.go

@@ -216,6 +216,27 @@ func TestONNXBaseline(t *testing.T) {
 func newBaselineRouter(t *testing.T, emb router.Embedder, llmR *router.LLMRouter) *router.Router {
 	t.Helper()
 	acts := router.DefaultActMatcher{Fns: actFns}
+	cls := newBaselineClassifier(t, emb)
+	return router.New(router.Config{
+		Grammars:   baselineGrammars(acts),
+		Classifier: cls,
+		Extractor: router.Extractor{
+			Time:  router.StubDateTimeParser{},
+			Acts:  acts,
+			Facts: router.DefaultFactParser{},
+		},
+		// The deployed gate, not a test-local one: a fixture scored at a looser
+		// threshold reports an accuracy no real turn would see.
+		Threshold: config.DefaultRouterThreshold,
+		LLM:       llmR,
+	})
+}
+
+// newBaselineClassifier — the seeded nearest-centroid classifier the cascade
+// runs. Split out of newBaselineRouter so the claim measurement can ask it for
+// its full ranking, not just the winner the Decision carries.
+func newBaselineClassifier(t *testing.T, emb router.Embedder) *router.Classifier {
+	t.Helper()
 	cls := router.NewClassifier(emb)
 	ctx := context.Background()
 	seeds := seedsWithIntent(t)
@@ -231,6 +252,15 @@ func newBaselineRouter(t *testing.T, emb router.Embedder, llmR *router.LLMRouter
 			t.Fatalf("seed %q: %v", text, err)
 		}
 	}
+	return cls
+}
+
+// baselineGrammars — the stage-0 rule set in the daemon's order (buildRouter in
+// cmd/mavend/voicewire.go). Split out of newBaselineRouter so the claim
+// measurement can run the same rules one at a time and see which of them
+// contend for the same utterance, which the cascade hides by stopping at the
+// first match.
+func baselineGrammars(acts router.ActMatcher) []router.Grammar {
 	grammars := router.DefaultGrammars(acts)
 	grammars = append(grammars, router.SystemTimeDateGrammars()...)
 	// Same order as buildRouter (voicewire.go). The fixture is only worth
@@ -248,19 +278,7 @@ func newBaselineRouter(t *testing.T, emb router.Embedder, llmR *router.LLMRouter
 	// "расскажи про X" is a world question the model called a fact, and the
 	// rule goes last because it matches on the first word alone (Vikunja #498).
 	grammars = append(grammars, router.NarrativeQueryGrammars()...)
-	return router.New(router.Config{
-		Grammars:   grammars,
-		Classifier: cls,
-		Extractor: router.Extractor{
-			Time:  router.StubDateTimeParser{},
-			Acts:  acts,
-			Facts: router.DefaultFactParser{},
-		},
-		// The deployed gate, not a test-local one: a fixture scored at a looser
-		// threshold reports an accuracy no real turn would see.
-		Threshold: config.DefaultRouterThreshold,
-		LLM:       llmR,
-	})
+	return grammars
 }
 
 // seedOrder — fixed iteration order over the corpus. Not cosmetic: a few