feat: Add support for renaming and retaining columns in data preprocessor

[dushyantbehl]

Description of the change

Add two flags to the data_config under dataset definition.

1. rename_columns which allows users to rename columns by passing a dict
2. retain_columns which allows users to specify which columns to retain and drop others.

This is different and outside the remove_columns argument to data handlers because people might want to use this functionality even without invoking data handlers.

This is especially needed in the case of interleaving multiple datasets with different features.

Related issue number

How to verify the PR

Was the PR tested

• I have added >=1 unit test(s) for every new method I have added.
• I have ensured all unit tests pass

[github-actions]

Thanks for making a pull request! 😃
One of the maintainers will review and advise on the next steps.

[willmj]

This looks good to me, this might require some investigation with iterable datasets after it is merged in.

[willmj]

Should we add one more test in tests/data/test_datapreprocessing.py to assert column names have been appropriately retained and renamed?

[dushyantbehl]

This looks good to me, this might require some investigation with iterable datasets after it is merged in.

I checked the Iterable datasets API and it does contain rename and retain APIs but you can add a few tests to your PR for sanity.

[dushyantbehl]

Should we add one more test in tests/data/test_datapreprocessing.py to assert column names have been appropriately retained and renamed?

@willmj added the tests here

[Abhishek-TAMU]

Apart for this small nit, LGTM! Thanks Dushyant!