Apply new input validation to cuml.metrics.cluster

[csadorf]

Apply new input validation to cuml.metrics.cluster.

Part of #7998

[coderabbitai]

Caution

Review failed

Failed to post review comments

📝 Walkthrough

Summary by CodeRabbit

• Refactor
  • Enhanced clustering metrics with improved input validation and more reliable data processing for adjusted Rand index, completeness score, entropy, homogeneity, mutual information, silhouette coefficient, and V-measure calculations to ensure more consistent behavior across different input formats.

Walkthrough

This PR refactors clustering metrics modules to replace legacy input-handling utilities with standardized NumPy-based validation (check_array, check_consistent_length) and updates pointer extraction to use the .data.ptr pattern across eight files in the cluster metrics module.

Changes

Clustering Metrics Validation Refactor

Layer / File(s)	Summary
Import and Utility Updates python/cuml/cuml/metrics/cluster/adjusted_rand_index.pyx, entropy.pyx, silhouette_score.pyx, utils.py	Removed legacy input_to_cuml_array and input_to_cupy_array imports; added check_array, check_consistent_length, and related validation utilities.
Data Validation Layer adjusted_rand_index.pyx, entropy.pyx, silhouette_score.pyx, utils.py	Inputs validated via check_array with explicit dtype enforcement (int32 for labels), 1D/shape assertion, and length consistency checks. Replaces previous input conversion paths with explicit validation before backend calls.
Label Preparation and Monotonic Mapping silhouette_score.pyx, utils.py	Monotonic label derivation reworked; uses cp.unique or make_monotonic with copy=True to produce new validated arrays. Class ranges computed from unique labels with lower bound fixed at 0.
Pointer Extraction Modernization completeness_score.pyx, homogeneity_score.pyx, mutual_info_score.pyx, silhouette_score.pyx, v_measure.pyx, adjusted_rand_index.pyx, entropy.pyx	Data pointers retrieved via .data.ptr instead of .ptr; updated across all files to align with new array object structure before C++ backend calls.
Backend Integration adjusted_rand_index.pyx, entropy.pyx, silhouette_score.pyx	Silhouette, ARI, and entropy backend calls updated to accept validated array pointers and shape parameters (n_rows, n_cols); sil_scores buffer handling adds NULL pointer support for optional inputs.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~50 minutes

Possibly related PRs

• rapidsai/cuml#8050: Applies the same check_array validation refactor to pairwise_kernels metrics.
• rapidsai/cuml#8044: Performs the same input_to_* to check_array migration pattern in regression metrics modules.
• rapidsai/cuml#8012: Migrates input_to_cupy_array to check_array validation utilities across additional metrics modules.

Suggested labels

Cython / Python, improvement, non-breaking

Suggested reviewers

• divyegala
• jcrist

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately and specifically describes the main change: applying new input validation to the cuml.metrics.cluster module, which aligns with the changeset across all modified metric files.
Description check	✅ Passed	The description is concisely related to the changeset, referencing the application of new input validation to cuml.metrics.cluster and linking to the parent issue #7998.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[jcrist]

/merge