The code was largely copied from splitters.py.

We now have code for the consecutive similarity splitter as well as the cumulative similarity splitter.