[BUG] QPD distribution and `CyclicBoosting` API non-compliance

[fkiraly]

@setoguchi-naoki, @FelixWick, I have to apologize in advance for this.

Due to an error in the test logic introduced in an update to class retrieval (see https://github.com/sktime/skpro/pull/189), probability distributions went uncovered for most of your PR's duration.

This is my fault for not noticing, and breaking it in the first place.

Bad news is that the QPD distribution have a few non-compliances:

• cdf method
• mean, var
• subsetting

CyclicBoosting as well:

• _predict_quantiles arg should be alpha, not quantiles

Good news is that this has not been released yet, and I was running more tests before the 2.2.0 release, which is what ultimately caught the problem.

I'll simply wait with 2.2.0 until we had time to fix this - happy to help.

For testing locally, you need to depend on the branch #189 until it is merged.

[fkiraly]

Mechanically, to remove failures from main while reeabling tests:

• I'll add skip commands for the failing estimators, see https://github.com/sktime/skpro/pull/193
• a PR fixing the remaining non-compliances should remove the skip commands to ensure testing

(the tests should still be runnable via check_estimators)

[FelixWick]

@setoguchi-naoki Can you give this a try, please? I can help if you face any issues, of course.

[fkiraly]

FYI, I made a branch were all tests are activated - you can use this for debugging, or branch off it for a fix branch: https://github.com/sktime/skpro/pull/194

I also attempted a naive fix to CyclicBoosting, renaming the argument: https://github.com/sktime/skpro/pull/195

[fkiraly]

For CyclicBoosting, renaming args seems to have worked: https://github.com/sktime/skpro/pull/195

So we now only need to worry about the QPD distributions.