Sometimes people ask why every little thing in the UI can't be configurable. This is why.
I don't see why. We have CONFIG in our software, our "framework" allows for something to either be in-the-personal-settings or not-in-the-personal-settings; it is the trivial to decide that it is "in or "out" or, indeed, to change our minds and move it from one to the other; it just needs the design to allow for this ability (which, given that there are "some" personal settings already it would be pretty unforgivable if that was not how the current stuff was designed). There might be a limitation on how much persistent memory/storage there is available though, or some other Gotcha..
the more toggles, the bigger your test matrix
We require no less than 14wks to finish and properly vet our code
I hope not on both count. I would expect there to be a fully automated software test system which was run regularly (probably daily (e.g. overnight) against a daily-build), that way the moment [well "by tomorrow"] that something is built the developer gets feedback as to whether there are any side-effects, and can adjust accordingly - rather than finding out, ages later, that there is a problem and then the fixes that are then made have major impacts on other parts of the system. That's how we used to do it, before automated-testing, and we could never predict a shipping date at all, let alone with any certainty. Machine-testing changed all that for us.
If there are any surprises, here, in testing of a formal QA release, which had not come to light in daily build testing, there would be blood on the carpet! OTOH users saying "I don't like that THIS does THAT" or "Its odd that if I do THIS then THAT displays" does arise (for us) in QA testing, but IMHO there is something seriously wrong with daily build / automated testing if the feedback from a QA Release is "
If I do THIS then THAT happens and THAT should never happen / is wrong".
That said we are seeing snafus during Tesla Production Releases, and we aren't party to how much, if any, stuff came up & was fixed during their QA Testing, so there is a lot of room for improvement.
The cost (both actual and PR) of a bug getting into a Production Release is so expensive that they must, surely??, be incentivised into preventing it five-9's, or more, of the time??