ENH: Styler.format_index() to display index values similarly to data-values with format()

[attack68]

• adds features and method with docstrings
• update user guide
• whats new
• add tests

This PR adds the method format_index so that the typical format routines can also be applied to display the index labels and columns headers of a Styler.

With all the recent functionality here is a complete example:

Notes

Also had to move _refactor_levels function from style.py to style_render.py for use here.

[jreback]

hmm why do we want to add methods like this? e.g. if you want to string format an index you can already

[attack68]

• Styler displays values without altering the underlying DataFrame. Using pd.Index.format returns a formatted Index which you need to overwrite as the DataFrames Index prior to calling Styler, therefore it breaks the typical usage pattern.
• pd.Index.format does not have the additional options available in format_index like HTML/LaTeX escaping, and has a different signature to Styler.format which might confuse users.
• Needing to overwrite the Index with your display format means you can lose functionality of indexing. Here is an example of how this helps preserve functionality:

Do throw this out to the team for feedback...

[attack68]

User Guide Updates:

[jreback]

give the allowed (e.g. 0,1,index,columns)

[attack68]

done

[attack68]

these tests consistently fail running the ipython in source/user_guide/scale.rst dealing with dask DataFrame.
But this PR is unrelated to that so I don't know the reason. Couldn't really find other PRs suffering from the same test failures?
The docs build locally without errors so its puzzling.

[jreback]

these tests consistently fail running the ipython in source/user_guide/scale.rst dealing with dask DataFrame.
But this PR is unrelated to that so I don't know the reason. Couldn't really find other PRs suffering from the same test failures?
The docs build locally without errors so its puzzling.

hmm, did dask version change?

[jreback]

looks like some actual doc failures: https://github.com/pandas-dev/pandas/pull/43101/checks?check_run_id=3570623205

[attack68]

looks like some actual doc failures: https://github.com/pandas-dev/pandas/pull/43101/checks?check_run_id=3570623205

everytime says build completed with 8 warnings, and every warning is in respect to /home/runner/work/pandas/pandas/doc/source/user_guide/scale.rst, which is an unaltered file, and never calls Styler or the impacted files here.

The docs build locally without warning so I am puzzled, will see if I can come up with anything

[attack68]

@jreback managed to get this to pass after merge. revert if necessary but I think should be ok now.

[jreback]

ok great. very odd that that was happening,