For compound transforms, there should be a setting that allows the user to see the intermediate columns

When the users use compound transforms, like `FeaturizeText`, the intermediate columns like: `*_TextNormalizer`, `*_TextNormalizer_WordTokenizer` etc are not part of the output schema.

I think it would be really useful for debugging and tunning purposes to be able to see how the text transforms step by step.

There should be an option to make those intermediate columns part of the output.