fix: resolve sklearn quality gate warnings

[egordm]

What does this PR do?

Resolves scikit-learn warning failures that prevented the full local quality gate from passing:

• Migrates the bundled splitter artifact by loading and reserializing it with the locked scikit-learn 1.9.1. The old and migrated artifacts produced bit-identical predictions on 260 deterministic inputs; the migrated artifact reloads without version warnings. The loading doctest and unit tests now exercise that clean load directly.
• Remove fractional bootstrap sampling from the tiny imputer test and example datasets, avoiding scikit-learn small-sample warnings.
• Keep the logistic-regression default as sklearn default rather than explicitly passing deprecated penalty="l2"; non-default penalties remain configurable without duplicating constructor setup.

This PR is stacked on #1135 and should be merged after it.

No linked issue.

Validation: uv run poe all --check passes locally. The focused splitter/doctest/logistic checks also pass (7 tests).

Type of change

• Bug fix
• New feature
• Breaking change (see checklist below)
• Documentation
• Refactor / chore / CI

Breaking changes checklist

• Public API, config schema, or serialized/pickled objects changed in a way that affects existing users
• If yes: see the breaking changes guide for the pickle/_migrate_state migration pattern

Migration path for existing users: N/A.

AI disclosure

• No AI assistance was used (beyond grammar/spelling)
• AI assistance was used — tool(s): GitHub Copilot
  • I have reviewed, understand, and can explain all AI-generated code in this PR
  • This is disclosed in a commit message (e.g. Assisted-by: <tool name>)

Checklist

• poe all --check passes locally
• Tests added/updated for the change
• Documentation updated (docstrings, user guide, examples) if needed
• Commits are signed off per our DCO (git commit -s)
• PR title follows Conventional Commits (e.g. feat: ..., fix: ..., feat!: ... for breaking changes)

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
75.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud