Describe the bug

Handling of infinite arrays when alpha is too small in the coverage calculation with the function regression_coverage_score().

To Reproduce
Steps to reproduce the behavior:

1. Go to notebook "ts-changepoint.ipynb"
2. Uncomment the following : "coverage_aci_pfit = regression_coverage_score(y_test, y_pis_aci_pfit[:, 0, 0], y_pis_aci_pfit[:, 1, 0])"
3. Launch all the notebook
4. See error

Expected behavior
How to handle alpha values in this case to calculate a regression coverage score without raising an error?

Screenshots

Desktop (please complete the following information):

• OS: IOS
• Browser : chrome
• MAPIE Version 0.8.6

Additional context
We get another type of error when allow_infinite_bounds = False in the .predict() method. It tells us that the number of samples must be greater than max(alpha, 1/alpha)