fix: OnStart ctx bug, rename session→user, full logout cleanup

- Use context.Background() for cleanup goroutine instead of
  OnStart ctx which is cancelled after startup completes
- Rename GetSessionCount→GetUserCount, DeleteStaleSessions→
  DeleteStaleUsers to reflect that sessions represent users
- HandleLogout now fully cleans up when last client disconnects:
  parts all channels (notifying members via QUIT), removes
  empty channels, and deletes the session/user record
- docker build passes, all tests green, 0 lint issues

internal/handlers/handlers.go

@@ -124,8 +124,16 @@ func (hdlr *Handlers) idleTimeout() time.Duration {
 	return dur
 }
 
-func (hdlr *Handlers) startCleanup(ctx context.Context) {
-	cleanupCtx, cancel := context.WithCancel(ctx)
+// startCleanup launches the idle-user cleanup goroutine.
+// We use context.Background rather than the OnStart ctx
+// because the OnStart context is startup-scoped and would
+// cancel the goroutine once all start hooks complete.
+//
+//nolint:contextcheck // intentional Background ctx
+func (hdlr *Handlers) startCleanup(_ context.Context) {
+	cleanupCtx, cancel := context.WithCancel(
+		context.Background(),
+	)
 	hdlr.cancelCleanup = cancel
 
 	go hdlr.cleanupLoop(cleanupCtx)
@@ -161,12 +169,12 @@ func (hdlr *Handlers) runCleanup(
 ) {
 	cutoff := time.Now().Add(-timeout)
 
-	deleted, err := hdlr.params.Database.DeleteStaleSessions(
+	deleted, err := hdlr.params.Database.DeleteStaleUsers(
 		ctx, cutoff,
 	)
 	if err != nil {
 		hdlr.log.Error(
-			"session cleanup failed", "error", err,
+			"user cleanup failed", "error", err,
 		)
 
 		return
@@ -174,7 +182,7 @@ func (hdlr *Handlers) runCleanup(
 
 	if deleted > 0 {
 		hdlr.log.Info(
-			"cleaned up stale clients",
+			"cleaned up stale users",
 			"deleted", deleted,
 		)
 	}