chore: fix misleading comment about CTAS and allow_append_only_skip (#19189)

fix misleading comment about CTAS and `allow_append_only_skip`

Previous comments about the behavior of CommitSink::allow_append_only_skip were incorrect. In fact,
CTAS operations are allowed to skip the update_table_meta operation inside CommitSink if no data
is inserted to the target table.

Author: dantengsky

src/query/storages/fuse/src/operations/common/processors/sink_commit.rs

@@ -386,10 +386,9 @@ where F: SnapshotGenerator + Send + Sync + 'static
             .is_some()
     }
 
-    /// Append-only inserts (e.g. `INSERT INTO t SELECT ...`) may skip committing if
-    /// nothing was written. Overwrite/CTAS (`CREATE OR REPLACE TABLE t AS SELECT ...`
-    /// or `INSERT OVERWRITE ...`) still need a snapshot even when nothing was written,
-    /// so we disable skipping when `AppendGenerator` is in overwrite mode.
+    /// Append-only inserts (e.g. `INSERT INTO t SELECT ...`) may skip committing if nothing was
+    /// written. `INSERT OVERWRITE ...` still need a snapshot even when nothing was written, so we
+    /// disable skipping when `AppendGenerator` is in overwrite mode.
     fn allow_append_only_skip(&self) -> bool {
         self.snapshot_gen
             .as_any()