249 clarification of the symmetry argument in cqr and more general documentation about cqr

[LacombeLouis]

Description

This change adds an example to the documentation that demonstrates the impact of the symmetry parameter in the MapieQuantileRegressor class. The example generates plots for prediction intervals with both symmetry=True and symmetry=False, allowing users to visualize and understand the differences.

Fixes #249.

Type of change

• New feature (non-breaking change which adds functionality)

How Has This Been Tested?

• Ran the script and visually inspected the generated plots to ensure they display correctly.
• Verified that the explanation accurately reflects the results shown in the plots.

Checklist

• I have read the contributing guidelines
• I have updated the HISTORY.rst and AUTHORS.rst files
• Linting passes successfully : make lint
• Typing passes successfully : make type-check
• Unit tests pass successfully : make tests
• Coverage is 100% : make coverage
• Documentation builds successfully : make doc

[thibaultcordier]

Super PR that clarifies the theoretical aspects of CQR. Bravo! I have made suggestions to add clarification.

Maybe rework to the notebook to improve it (add a legend, use a more meaningful example, show visually where quantile models and quantile corrections come into play in interval estimation).

[thibaultcordier]

Thanks for adding your comments to the notebooks.

To make sure we're talking about the same thing, I've made some suggestions to highlight the difference between quantile regressors with the symmetric/asymmetric option.

The code could not be put into words because CQRs are not random_state parameters. To check and see how we can have the same regressor for both.

Another possibility is to have a single MapieQuantileRegressor and change the attribute by hand during prediction (without re-fit the model). I think this is a preferable solution.

[thibaultcordier]

Here's my second proposal that I presented earlier to compare the symmetrical and asymmetrical options.

[thibaultcordier]

Ready to merge. Thank you for your final modifications.