[Feature] Add CatBoostCvExperiment integration (#229)

[coder-jayp]

Description

Adds a new CatBoostCvExperiment integration that allows Hyperactive to perform cross-validation directly using CatBoost's native catboost.cv() function.

This enables efficient hyperparameter tuning for CatBoost models with built-in CV.

Related Issues

Closes #229

Type of Change

• [ENH] - New feature (non-breaking change adding functionality)

How was this solved?

• Created new class CatBoostCvExperiment in src/hyperactive/experiment/integrations/catboost_cv.py
• Implemented _evaluate() to run catboost.cv() with fixed CV settings and dynamic hyperparameters
• Added optional metric and loss_function parameters (with automatic custom_metric handling when they differ)
• Set direction tag to "lower" (CatBoost objectives are minimization)
• Added CatBoostCvExperiment to init.py exports
• Added two pytest tests in src/hyperactive/tests/test_integrations/test_catboost_cv.py:
  • Basic Logloss run
  • AUC metric with Logloss objective (verifies metric/loss separation)

Checklist

• PR title includes appropriate tag: [BUG], [ENH], [DOC] or [MNT]
• Linked to related issue (if applicable)
• Code passes make check (lint, format, isort)
• Tests added/updated for changes (if applicable)
• Documentation updated (if applicable)

Testing

Run the integration tests (verbose mode recommended):

pytest src/hyperactive/tests/test_integrations/test_catboost_cv.py -v