[ENH] added update capability in GridSearchCV and RandomizedSearchCV

[patelchaitany]

Reference Issues/PRs

#1050

What does this implement/fix? Explain your changes.

This add the update capability for GridSearchCV and RandomizedSearchCV

BaseGridSearch:

• Added update_behaviour="full_refit" parameter to init
• Added "capability:update" to tags_to_clone so the tag propagates from the inner estimator
• Added self._X, self._y, self._C storage at end of _fit() for data accumulation
• Added _update() method with three strategies: "full_refit", "inner_only", "no_update"
• Added _update_data() helper for concatenating old + new data

GridSearchCV:

• Added update_behaviour parameter to init, passed through to super
• Added update_behaviour docstring entry
• Added test params with OnlineRefit-wrapped estimator

RandomizedSearchCV:

• Same three changes as GridSearchCV (parameter, docstring, test params)

Does your contribution introduce a new dependency? If yes, which one?

No

What should a reviewer concentrate their feedback on?

Did you add any tests for the change?

Any other comments?

PR checklist

For all contributions

• I've added myself to the list of contributors with any new badges I've earned :-)
  How to: add yourself to the all-contributors file in the skpro root directory (not the CONTRIBUTORS.md). Common badges: code - fixing a bug, or adding code logic. doc - writing or improving documentation or docstrings. bug - reporting or diagnosing a bug (get this plus code if you also fixed the bug in the PR).maintenance - CI, test framework, release.
  See here for full badge reference
• The PR title starts with either [ENH], [MNT], [DOC], or [BUG]. [BUG] - bugfix, [MNT] - CI, test framework, [ENH] - adding or improving code, [DOC] - writing or improving documentation or docstrings.

For new estimators

• I've added the estimator to the API reference - in docs/source/api_reference/taskname.rst, follow the pattern.
• I've added one or more illustrative usage examples to the docstring, in a pydocstyle compliant Examples section.
• If the estimator relies on a soft dependency, I've set the python_dependencies tag and ensured
  dependency isolation, see the estimator dependencies guide.

[fkiraly]

Thanks, good start!

Some change requests:

• please do not store the data by default, this causes memory leaks in current code. Instead, make the default option the current one - no update happens - and only store data if a strategy is chosen that requires this.
• No pull request is ready without docstrings. Please update the docstrings properly.

[patelchaitany]

Thanks for the review - both points are addressed

[fkiraly]

Nice!

I think the tag capability:update is not correctly set though?

If full_refit, it should be True; if no_update, it should be False, If inner_only, it should be clone the tag of the inner estimator.

Currently, it is always cloned.

The class tag (in _tags) should be True, since for some setting of parameters the capability is True.

[patelchaitany]

Thanks @fkiraly for catching this. Sorry I overlooked the capability:update tagging in the first pass.

I have pushed a fix that sets the class tag to True in _tags and sets the instance tag from update_behaviour: False for no_update, True for full_refit, and clones from the inner estimator for inner_only.

Happy to adjust if anything else should change.

[fkiraly]

Nice!