Potential Han model error

[ekiefl]

Some errors in the Han implementation in pooltool were discovered, and I wanted to flag them here in case your implementation also shares these errors. The missing mu is apparently an error in the paper itself.

ekiefl/pooltool#247

[tailuge]

Thanks, I'll take a look - Han has been in there such a long time, did the fixes show any meaningful change in behaviour?

[tailuge]

I've made some recent updates, I'm not using Han for cushions so won't bother changing that for now. I'm still interested is best values for mathaven constants if your users researched that. I put a bot in, it was missing some pots and it led me to a fix on collision resolution, that's my main change.