Add save/load methods to anomaly detection classes

[robrui]

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 #3051.

Summary

Adds save() and load() methods to the four anomaly detection base classes (AnomalyScorer, Detector, Aggregator, AnomalyModel). This makes it possible to persist trained scorers like KMeansScorer, detectors like QuantileDetector, aggregators, and anomaly models to disk and reload them later without retraining. Uses the same pickle-based approach as ForecastingModel.save/load.

Other Information

All subclasses inherit save/load automatically. The full anomaly detection test suite passes (229 tests).

[codecov]

Codecov Report

❌ Patch coverage is 80.51948% with 15 lines in your changes missing coverage. Please review.
✅ Project coverage is 96.41%. Comparing base (b0ccd3a) to head (a89f247).

Files with missing lines	Patch %	Lines
darts/ad/aggregators/aggregators.py	78.94%	4 Missing ⚠️
darts/ad/anomaly_model/anomaly_model.py	78.94%	4 Missing ⚠️
darts/ad/detectors/detectors.py	78.94%	4 Missing ⚠️
darts/ad/scorers/scorers.py	85.00%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3129      +/-   ##
==========================================
- Coverage   96.55%   96.41%   -0.14%     
==========================================
  Files         160      160              
  Lines       17285    17362      +77     
==========================================
+ Hits        16689    16740      +51     
- Misses        596      622      +26     

☔ View full report in Codecov by Harness.
📢 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.