feat: allow stride=forecast_horizon for anomaly forecasting model

[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 #2673.

Summary

• Expose the stride argument in ForecastingAnomalyModel.predict_series, it must be equal to 1 or forecast horizon
• Add the corresponding tests

Other Information

Using stride=forecast_horizon should speed up the forecasts at the cost of accuracy, not sure if we should also test this aspect?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.12%. Comparing base (19b17d2) to head (395e077).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2709      +/-   ##
==========================================
- Coverage   94.18%   94.12%   -0.07%     
==========================================
  Files         141      141              
  Lines       15598    15605       +7     
==========================================
- Hits        14691    14688       -3     
- Misses        907      917      +10     

☔ 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]

Thanks @madtoinou for this PR. However, I'm not sure that it fully solves the issue #2673. The issue was that it's not possible to fit() and score() with a ForecastingAnomalyModel with a stride != 1 (but == horizon).