fix(startup): sweep open scrobbles before accepting playback

The open-scrobble sweep was queued in the deferred background-init list,
which runs concurrently with the HTTP listener; a resume immediately after
restart could start new scrobbles before the previous process's open
sessions were stopped, leaving overlapping/stale scrobbles on remote
providers. Run the sweep synchronously before the listener starts, bounded
by a 30s timeout so an unreachable provider can't hang startup (the heavier
non-critical init stays deferred).

Addresses PR #21 review (P2).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

cmd/silo/main.go

@@ -1023,11 +1023,6 @@ func main() {
 			WithMatcher(historyimport.NewMatcher(historyRepo)).
 			WithWatchState(watchstate.NewService(userStoreProvider).WithStableIdentityResolver(historyIdentity)).
 			WithUserStoreProvider(userStoreProvider)
-		backgroundInit = append(backgroundInit, func(ctx context.Context) {
-			if err := watchProviderService.SweepOpenScrobbles(ctx); err != nil {
-				slog.Warn("failed to sweep open watch provider scrobbles", "error", err)
-			}
-		})
 	}
 	deps.SessionMgr = sessionMgr
 	deps.PlaybackRealtimeHub = playback.NewRealtimeHub()
@@ -1683,6 +1678,19 @@ func main() {
 		}()
 	}
 
+	// Stop scrobble sessions left open by the previous process BEFORE the
+	// listener accepts playback requests, so a resume immediately after a
+	// restart doesn't create overlapping/stale scrobbles on remote providers.
+	// Bounded by a timeout so an unreachable provider can't hang startup (which
+	// is why the rest of non-critical init stays deferred below).
+	if watchProviderService != nil {
+		sweepCtx, cancelSweep := context.WithTimeout(appCtx, 30*time.Second)
+		if err := watchProviderService.SweepOpenScrobbles(sweepCtx); err != nil {
+			slog.Warn("failed to sweep open watch provider scrobbles", "error", err)
+		}
+		cancelSweep()
+	}
+
 	errCh := make(chan error, 2)
 	go func() {
 		slog.Info("HTTP server listening", "addr", cfg.Server.Listen)