Describe the workflow you want to enable

DP-Means is a nonparametric version of K-Means, which is rooted in the Dirichlet Process Mixture Model.
Proposed by Kulis and Jordan in 2011, it is already a well cited and established algorithm.

Recently, we have published a version of it which scales very good, comparable to K-Means and is equivalent to the original algorithm (https://openreview.net/pdf?id=rnzVBD8jqlq).

Enabling it in scikit-learn would benefit the community, and would be great in my opinion.

Describe your proposed solution

We have already utilized scikit-learn codebase in order to implement our algorithm efficiently, (https://github.com/BGU-CS-VIL/pdc-dp-means/tree/main/cluster), so I can easily fix this into a pull-request to scikit-learn code base (will need to add the relevant tests and docs as well).

As such, if this feature request is approved, I will work on the relevant pull request in order to add this to the code base.

Describe alternatives you've considered, if relevant

No response

Additional context

No response