[ENH] Allow string expressions for second_level_contrast of non_parametric_inference

[ymzayek]

• Closes [BUG] second-level contrasts in string format in non_parametric_permutation() #3807, closes [BUG] Trouble running arithmetic in second_level_contrast of non_parametric_inference #4140

TODO:

• Add test

[github-actions]

👋 @ymzayek Thanks for creating a PR!

Until this PR is ready for review, you can include the [WIP] tag in its title, or leave it as a github draft.

Please make sure it is compliant with our contributing guidelines. In particular, be sure it checks the boxes listed below.

• PR has an interpretable title.
• PR links to Github issue with mention Closes #XXXX (see our documentation on PR structure)
• Code is PEP8-compliant (see our documentation on coding style)
• Changelog or what's new entry in doc/changes/latest.rst (see our documentation on PR structure)

For new features:

• There is at least one unit test per new function / class (see our documentation on testing)
• The new feature is demoed in at least one relevant example.

For bug fixes:

• There is at least one test that would fail under the original bug conditions.

We will review it as quick as possible, feel free to ping us with questions if needed.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (f4b03d4) 91.69% compared to head (834235c) 91.91%.
Report is 29 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4150      +/-   ##
==========================================
+ Coverage   91.69%   91.91%   +0.21%     
==========================================
  Files         144      145       +1     
  Lines       16150    16324     +174     
  Branches     3359     3404      +45     
==========================================
+ Hits        14809    15004     +195     
+ Misses        796      777      -19     
+ Partials      545      543       -2     

Flag	Coverage Δ
macos-latest_3.10_test_plotting	91.69% <100.00%> (?)
macos-latest_3.11	?
macos-latest_3.11_test_plotting	91.69% <100.00%> (?)
macos-latest_3.12	?
macos-latest_3.12_test_plotting	91.69% <100.00%> (?)
macos-latest_3.8_test_plotting	91.65% <100.00%> (?)
macos-latest_3.9	?
macos-latest_3.9_test_plotting	91.65% <100.00%> (?)
ubuntu-latest_3.10	?
ubuntu-latest_3.10_test_plotting	91.69% <100.00%> (?)
ubuntu-latest_3.11	?
ubuntu-latest_3.11_test_plotting	91.69% <100.00%> (?)
ubuntu-latest_3.12	?
ubuntu-latest_3.12_test_plotting	91.69% <100.00%> (?)
ubuntu-latest_3.12_test_pre	91.69% <100.00%> (?)
ubuntu-latest_3.8_test_min	68.83% <100.00%> (?)
ubuntu-latest_3.8_test_plot_min	91.41% <100.00%> (?)
ubuntu-latest_3.8_test_plotting	91.65% <100.00%> (?)
ubuntu-latest_3.9	?
ubuntu-latest_3.9_test_plotting	91.65% <100.00%> (?)
windows-latest_3.10	?
windows-latest_3.10_test_plotting	91.66% <100.00%> (?)
windows-latest_3.11	?
windows-latest_3.11_test_plotting	91.66% <100.00%> (?)
windows-latest_3.12	?
windows-latest_3.12_test_plotting	91.66% <100.00%> (?)
windows-latest_3.8_test_plotting	91.62% <100.00%> (?)
windows-latest_3.9	?
windows-latest_3.9_test_plotting	91.63% <100.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[Remi-Gau]

So far this looks good to me.

Not sure what was causing me issue with this.

[Remi-Gau]

@bthirion
may be worth checking if we can add tests for some of the edge cases your were mentioning.

[ymzayek]

So far this looks good to me.

Not sure what was causing me issue with this.

It took me a while to also come up with a rather simple solution which is why I wasn't confident about it 😅. I was initially was focused on having to change things in permuted_ols but if this makes sense to you and others I don't see a problem with allowing it

[bthirion]

All changes LGTM and I don't see whay non binary contrasts would raise issues. This code is better than I thought ;-)