operator: Move AllZero check after StatefulSet deletion during teardown

When a StretchCluster was being torn down, the AllZero() early return
at the top of reconcileDecommission exited before reaching the
ToDelete() loop. StatefulSets with zero replicas were never deleted,
which blocked their owning NodePool custom resources from being
garbage collected.

Move the AllZero check after the scale-down and StatefulSet deletion
logic so that zero-replica sets are cleaned up before the function
returns.

Author: RafalKorepta

operator/internal/controller/redpanda/nodepool_controller.go

@@ -213,6 +213,7 @@ func (r *NodePoolReconciler) Reconcile(ctx context.Context, req mcreconcile.Requ
 
 	// Examine if the object is under deletion
 	if !pool.ObjectMeta.DeletionTimestamp.IsZero() {
+		logger.V(log.TraceLevel).Info("deleting finalizer")
 		if controllerutil.RemoveFinalizer(pool, FinalizerKey) {
 			if err := k8sClient.Update(ctx, pool); err != nil {
 				logger.Error(err, "updating cluster finalizer")
@@ -228,6 +229,7 @@ func (r *NodePoolReconciler) Reconcile(ctx context.Context, req mcreconcile.Requ
 	// If any changes are made, persist the changes and immediately requeue to
 	// prevent any cache / resource version synchronization issues.
 	if controllerutil.AddFinalizer(pool, FinalizerKey) || feature.SetDefaults(ctx, feature.V2Flags, pool) {
+		logger.V(log.TraceLevel).Info("adding finalizer")
 		if err := k8sClient.Update(ctx, pool); err != nil {
 			logger.Error(err, "updating cluster finalizer or Annotation")
 			return ignoreConflict(err)