Skip duplicate-column concat test when pyarrow is not installed

The non-pyarrow unit test jobs import pyarrow as None, so the new test
failed with AttributeError instead of being skipped like its neighbors.

Co-authored-by: Isaac <no-reply@databricks.com>
Signed-off-by: Mehmet Can Kurt <mehmet.cankurt@databricks.com>

Author: mehmetcankurt

tests/unit/test_util.py

@@ -174,6 +174,7 @@ def old_concat(tables, promote_options=None, **kwargs):
         result_table = concat_table_chunks([arrow_table1, arrow_table2])
         assert result_table.column("col1").to_pylist() == [1, 2, 3, 4]
 
+    @pytest.mark.skipif(pyarrow is None, reason="PyArrow is not installed")
     def test_concat_table_chunks_duplicate_column_names(self):
         # Duplicate column names are valid SQL (e.g. SELECT 1 AS a, 2 AS a).
         # Schema promotion rejects them with ArrowInvalid, so when chunk