[WIP] Benchmark Script

[NivekT]

Stack from ghstack:

• -> #734

This PR is primarily focused on adding more datasets for benchmarking.

Notable changes that are in progress:

• Using PrototypeMultiprocessingReadingService as that will become the official one in the near future
• Adding various datasets to benchmarking script
  • [WIP] Need to modify make_dp and etc to make sure that works for each dataset (I am thinking that I can take certain implementation steps from the prototype folder in vision or preferably directly use those)
  • [WIP] Need to add some datasets to the cluster's ontap
• Add support for the equivalent datasets for external cluster