WIP: DRAFT: E-graphs in the PMC for normalized refinement types

WIP, started over at a new MR... I think things are much clearer now, but I still need to figure out a way to handle positive Constructor info