Concurrent Apply can lead to child node applies before parent, so
undo/redo operations will perform. This leads to performance degradation
in case of tree with many sublevels.

Signed-off-by: Dmitrii Stepanov <d.stepanov@yadro.com>

@@ -384,7 +376,7 @@ func (s *Service) syncLoop(ctx context.Context) {
 			return
 		case <-s.syncChan:
 			ctx, span := tracing.StartSpanFromContext(ctx, "TreeService.sync")
-			s.log.Debug(logs.TreeSyncingTrees)

@@ -202,3 +190,1 @@
-				unappliedOperationHeight = min(unappliedOperationHeight, m.Time)
-				heightMtx.Unlock()
-				return err
+		if len(batch) == 1000 {

@@ -561,0 +609,4 @@
+	}
+
+	ops := make([]*Move, 0, len(m))
+	for _, op := range m {

@@ -106,6 +106,33 @@ func (s *Shard) TreeApply(ctx context.Context, cnr cidSDK.ID, treeID string, m *
 	return s.pilorama.TreeApply(ctx, cnr, treeID, m, backgroundSync)
 }
 
+// TreeApply implements the pilorama.Forest interface.

@@ -561,0 +620,4 @@
+			var logKey [8]byte
+			binary.BigEndian.PutUint64(logKey[:], op.Time)
+			seen := b.Get(logKey[:]) != nil
+			if !seen {