Add better initialisation methods

[tecosaur]

As far as I can tell, only kmeans++ is currently implemented. Looking at https://www.mdpi.com/1999-4893/14/1/6 "Improving Scalable K-Means++" it looks like SRPK-means‖ could be a good method to have available 🙂.

[PyDataBlog]

@tecosaur thanks for the suggestion. It will be added to list of things to do. We'll keep this open until it is implemented.