Support remote and cluster table functions for parallel_distributed_insert_select

Why it matters

This enhancement enables users to perform distributed insert-select operations in parallel when using remote() and cluster() table functions. It improves efficiency and scalability for distributed data insertion tasks by leveraging parallelism across remote servers or clusters.

How to use it

Users can now utilize remote() and cluster() table functions inside parallel_distributed_insert_select queries to execute distributed insert-select operations in parallel. No additional configuration is needed beyond existing usage patterns for these functions.