Major Description
We have many tests with operations on different Series in spark/python/pyspark/pandas/tests/data_type_ops/ that assume the result's index to be sorted and then compare to the pandas' behavior. The assumption is wrong, so we should sort index of such result before comparing with pandas.
