I remember building something very similar (actually it was more or less identical) using Mathematica many, many years ago. I never released it, because it was a fringe project with fringe dependencies and an even tinier audience back then. Oh, and it was terribly slow, because GPU acceleration was not yet really a thing. So it's kind of sad to see that this project here is using MATLAB, because that drastically cuts down its own potential audience as well. If I was developing something like that again today, I'd do it in Python/CUDA or at least in a full OSS approach. Over the years there have been a bunch of publications that came up with interesting geometries with curious properties, but everyone had to re-do the analysis software from scratch. Something truly open and maintainable might improve the situation and incentivise more work in that direction.