Support for sklearn 1.6 conformance testing

[ethanglaser]

Description

• scikit-learn-intelex PR to support internal conformance checks on sklearn 1.6 for CPU and 1.4 for GPU
• deselected tests updates to handle xfail/xpass discrepancies and a handful of outlying tests - all were investigated and large-scale failures have been fixed
• updates to daal4py logistic_path logic to align with scikit-learn
• updates to ridge to align coefficient and intercept shapes with scikit-learn logic

---

PR should start as a draft, then move to ready for review state after CI is passed and all applicable checkboxes are closed.
This approach ensures that reviewers don't spend extra time asking for regular requirements.

You can remove a checkbox as not applicable only if it doesn't relate to this PR in any way.
For example, PR with docs update doesn't require checkboxes for performance while PR with any change in actual code should have checkboxes and justify how this code change is expected to affect performance (or justification should be self-evident).

Checklist to comply with before moving PR from draft:

PR completeness and readability

• I have reviewed my changes thoroughly before submitting this pull request.
• I have commented my code, particularly in hard-to-understand areas.
• I have updated the documentation to reflect the changes or created a separate PR with update and provided its number in the description, if necessary.
• Git commit message contains an appropriate signed-off-by string (see CONTRIBUTING.md for details).
• I have added a respective label(s) to PR if I have a permission for that.
• I have resolved any merge conflicts that might occur with the base branch.

Testing

• I have run it locally and tested the changes extensively.
• All CI jobs are green or I have provided justification why they aren't.
• I have extended testing suite if new functionality was introduced in this PR.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Flag	Coverage Δ
azure	79.83% <85.71%> (+0.07%)	⬆️
github	73.56% <42.85%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
onedal/linear_model/linear_model.py	83.58% <ø> (-0.36%)	⬇️
sklearnex/linear_model/ridge.py	82.46% <100.00%> (+4.23%)	⬆️

🚀 New features to boost your workflow:

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

[ethanglaser]

http://intel-ci.intel.com/f02f4045-991c-f196-89ba-a4bf010d0e2d

[yuejiaointel]

Any impact of these extra checks on throughput or memory?

[ethanglaser]

Any impact of these extra checks on throughput or memory?

Generally just additions of sklearn version checks so should have negligible impact.

Also for reference, there are roughly 3000 new tests running in 1.6.1 compared to 1.5.2, can yield 10-20% increase in runtimes for these steps (which amounts to a minute or two of runtime).

[icfaust]

Relatively straightforward, just some small questions/changes. I think we can get this easily ready for merge for when private CI is available again. The only actual change is to duplicate the Ridge changes to LinearRegression and the incrementals.

[icfaust]

Could you do this for Linear Regression and the related incremental algos so that they all behave the same/ reduce maintenance?

[ethanglaser]

I am not sure its relevant for the others, it originates from scikit-learn/scikit-learn#19746 which specifically addresses Ridge

[icfaust]

https://github.com/scikit-learn/scikit-learn/blob/main/sklearn/linear_model/tests/test_common.py#L197 Is the answer hidden within the linked PR. Place notes into LinearRegression and IncrementalLinearRegression about this deviation, and add this change in return values to IncrementalRidge.

[icfaust]

@ethanglaser ping me directly once changes are made, so we can run private CI and merge this quickly.