[Feature Request]: Add Machine Learning capabilities to Apache Hop

### What would you like to happen?

I have been exploring the Stanford NLP transform in Apache Hop and was wondering whether there are plans or existing work around more general Machine Learning capabilities in Hop.

For example, capabilities such as data preprocessing, model training, model saving/loading and making predictions could be useful for use cases such as text classification or anomaly detection.

Is there already any work in progress in this area, or would this be something the community would consider adding to Apache Hop?

Thank you for your feedback.

### Issue Priority

Priority: 2

### Issue Component

Component: Transforms

Start by reviewing the existing Stanford NLP transform and any related work in Apache Hop. Clarify the desired scope across preprocessing, model training, model saving/loading, and predictions; the work is done only when the community agrees on a concrete capability and implementation boundary.

Written by the indexing model from the issue text.