ENH: Styler.format_index split PR base components

[attack68]

Splits #43101 to debug failing tests

All this PR does is add a method and fix failing tests because of a minor change to jinja2 templates.

[attack68]

This PR suffers the same undiagnosable fault as the mentioned PR..
If anyone fancies having a crack at solving the mysterious failure, please shout out.

[jreback]

try renaming this to format_index2 as a test and see if it works

[attack68]

try renaming this to format_index2 as a test and see if it works

so that worked? whats the scoop?

[jreback]

try renaming this to format_index2 as a test and see if it works

so that worked? whats the scoop?

wow. so was thinking that .format_index shadows something, but we don't have a method by that name so that's obviously bunk, but maybe we somehow autogenerate something.

[attack68]

try renaming this to format_index2 as a test and see if it works

so that worked? whats the scoop?

wow. so was thinking that .format_index shadows something, but we don't have a method by that name so that's obviously bunk, but maybe we somehow autogenerate something.

so you wanna keep the method named like this or something else. i've not faced issue like this before so bit clueless.

[jreback]

must be some sphinx thing

ideally let's keep format_index though could call format_axis

[attack68]

OK well this PR now uses format_axis and the original PR still has format_index, which I prefer since it complements apply_index and applymap_index, downside being it breaks the CI and I can't fix it! So you can choose either! (I'll fix up the tests and add the doc examples if you go with this one)

[jreback]

yeah might try to google this
i suspect we can set some sphinx option in conf.py to avoid this