Basically, it was publicly tested for eight months, the response was favorable, and nobody found any problems with it in any levels. (Except for with speed destruct scenery, and that was fixed prior to release.) The proposal for three months prior to release was that specific levels should be able to turn it off, which they can, and nobody objected to that proposal. This wasn't on our radar as something that might have issues--as far as we could tell, this was something that the community looked at for months and seemed to like.
It does, though so far everyone seems on board with the proposal to turn /quirks into a server-side test level thing,. Once that happens it should definitely turn the slopes change off, among various other changes... I'm not sure yet if that means, when it's a test level, ignore the value of jjPlayersStickToSlopes? Or set it to false upon the gamemode changing? Exact implementation details tbd, there's probably a right answer we just haven't thought it through yet.
So I'm not sure what this refers to, but the absolute best way to observe potential physics differences between versions is .j2m files: if you have one of those, that would be great to look at