That's a wall of text that doesn't make much sense from my quick skimming so I do not want to read in detail, and basically my point is that things like Simex and others do not have any preference to what the "desired" output any engine may produce, they just compare results in between each other at scale and show how much they differ. In which case, it becomes obvious that it is "just Stockfish", even lesser different than some older Stockfish itself and you don't seem to deny that.
While in every of your mentioned test it is nothing more than a cherry pick. And indeed if one goes for a chosen list of "puzzles" to see how well engine solves, then there are engines like Crystal which does specifically that(where most of code in ShashinChess was yoinked from btw) and it still doesn't have anything to do with Shashin theory. Your explaintion sounds more like Andrea Manzo just put those Crystal code under a tickbox and say "this is Shashin theory". I think for a genuine implementation one needs do more than giving nicknames to some code they didn't write.
Shashin theory
Moderator: Ras