Artifact cleanup

[natoverse]

Cleans up a large number of duplicated or unused fields in the output artifacts, and renames a few things for consistency.

[jgbradley1]

Now that the column names have been cleaned up quite a bit, is this a good PR to add docstrings to each workflow step to document the expected schema of the input/output dataframes?

As an alternative approach to docstrings, what do you think about using pydantic models to reference and validate the schema of each dataframe?

[natoverse]

Now that the column names have been cleaned up quite a bit, is this a good PR to add docstrings to each workflow step to document the expected schema of the input/output dataframes?

As an alternative approach to docstrings, what do you think about using pydantic models to reference and validate the schema of each dataframe?

I would much prefer Pydantic models, as I've noticed some of those docstrings were out of date anyway. We can add that at any time though I think. FWIW, I also added a docs page for the parquet schemas to help answer user questions.