fix(distributed): track in-flight for SoundDetection requests (#10475)

The distributed router wraps backend clients in InFlightTrackingClient so
the eviction logic knows which replicas are actively serving. Every
inference method must be wrapped: track() increments in-flight on entry
and decrements (plus fires onFirstComplete, which releases the load-time
reservation) on return.

SoundDetection was added after the tracking client and never got a
wrapper, so its calls fell through to the embedded passthrough Backend.
The increment/decrement never ran and, critically, onFirstComplete never
fired, so the reservation set at model load was never released - leaving
in-flight stuck at 1 and the replica permanently ineligible for eviction.

Wrap SoundDetection like the other non-LLM methods and cover it in the
"non-LLM inference methods track in-flight" table test.


Assisted-by: Claude:claude-opus-4-8 [Claude Code]

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>

core/services/nodes/inflight.go

@@ -218,6 +218,12 @@ func (c *InFlightTrackingClient) Score(ctx context.Context, in *pb.ScoreRequest,
 	return res, c.reconcile(err)
 }
 
+func (c *InFlightTrackingClient) SoundDetection(ctx context.Context, in *pb.SoundDetectionRequest, opts ...ggrpc.CallOption) (*pb.SoundDetectionResponse, error) {
+	defer c.track(ctx)()
+	res, err := c.Backend.SoundDetection(ctx, in, opts...)
+	return res, c.reconcile(err)
+}
+
 func (c *InFlightTrackingClient) AudioEncode(ctx context.Context, in *pb.AudioEncodeRequest, opts ...ggrpc.CallOption) (*pb.AudioEncodeResult, error) {
 	defer c.track(ctx)()
 	res, err := c.Backend.AudioEncode(ctx, in, opts...)

core/services/nodes/inflight_test.go

@@ -408,6 +408,13 @@ var _ = Describe("InFlightTrackingClient", func() {
 				return err
 			})
 		})
+
+		It("SoundDetection", func() {
+			assertTracked(func() error {
+				_, err := client.SoundDetection(context.Background(), &pb.SoundDetectionRequest{})
+				return err
+			})
+		})
 	})
 
 	Describe("stale model reload (self-heal)", func() {