[FEA]: Support newer XGBoost model formats

[cwharris]

Is this a new feature, an improvement, or a change to existing functionality?

New Feature

How would you describe the priority of this feature request

High

Please provide a clear description of problem this feature solves

With the upgrade to RAPIDS 24.10, cuml has changed the default model_type argument for cuml.ForestInference.load from xgboost to the newer universal binary json format. To account for this, the RAPIDS upgrade PR will automatically pass model_type=xgboost to keep tests passing. However, this means users still won't be able to use the newer model formats (ubj and json) with ClassificationStage.

Describe your ideal solution

Update ClassificationStage (gnn-fraud-classification) to support a model_type argument which defaults to None (or accepts a kwarg for the argument) so we pick up the default cuml model_type argument.

Update all xgboost models in the Morpheus repositories to use the new ubj format by loading and re-saving the models.

Update invocations of ClassificationStage to account for the updated models and model_type argument - this may not be necessary after updated the models, assuming they match the default type.

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct
• I have searched the open feature requests and have found no duplicates for this feature request