feat: adding support for getting estimator based on quantile

[madtoinou]

Checklist before merging this PR:

• Mentioned all issues that this PR fixes or addresses.
• Summarized the updates of this PR under Summary.
• Added an entry under Unreleased in the Changelog.

Fixes #2596.

Summary

• added logic to retrieve estimator based on quantile in addition to the horizon and target_dim
• added the corresponding tests

Other Information

• The interactions between get_multioutput_regressor and get_estimator is a bit ambiguous, we could maybe simplify it.
• Interestingly, it seems to be possible to pass both quantiles=[...] with likelihood="poisson" without raising an exception. We abuse this mechanism in the tests but it should probably not be possible in the first place...

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.10%. Comparing base (fabba0a) to head (b5ce31e).
Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2716      +/-   ##
==========================================
- Coverage   94.18%   94.10%   -0.08%     
==========================================
  Files         141      141              
  Lines       15591    15596       +5     
==========================================
- Hits        14684    14677       -7     
- Misses        907      919      +12     

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

🚀 New features to boost your workflow:

• ❄ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[dennisbader]

Looks great, thanks a lot @madtoinou

As discussed, let's let get_estimator handle all the logic and remove get_multioutput_estimators.

Then all good to be merged :)

[dennisbader]

Perfect 💯