Yes, I understand that. If we just say, "rulings over rules.", though, every time something comes up, there'd be precious little to discuss about 5e. You can make a ruling about pretty much any issue that you encounter and just move on.
Quite true. That's a feature, not a bug. If it happens once, you make a ruling, and go on. If it happens a lot at your table, you make a house rule. If we find it is a common occurrence across many tables, then we start considering that the system has a notable fault that needs correcting.
This thread is specifically about cantrips and the issues with them being at-will, so that's what I'm discussing.
You appear to be focused on what seems to be an edge case, that can happen in theory, but in practice is pretty rare.
There's two points to make about changing rules:
1) Any rules design choice has consequences. If you aren't in a position to do a lot of playtesting, you are apt to miss some of the unintended consequences, and introduce a problem that's worse than the thing you were trying to fix.
2) If you are changing rules that are central to some operations, you are apt to be "moving someone's cheese" as they may say in the business world - changing something that someone *really* liked.
These are not so great things that GMs perhaps should try to avoid, especially when the issue you raise can probably be handled with
a conversation, rather than a rule....
"Hey, folks? I'd like to try to preserve the spirit of these rules, that make sure that folks don't have to sit around doing nothing in combat because they are out of spells or other special maneuvers to use. But, there are some technical edge cases. Could you, you know, *not* try to abuse those edge cases? Otherwise, I might have to consider some exhaustion rules or something like that, which may otherwise cramp your style..."
And suddenly the problem goes away, because breaking down castle walls with cantrips probably wasn't central to anyone's character concept, so they can just not try to do that, and preserve all the stuff they do like. You don't get anyone spamming basic attacks for hours, they get not having to worry about *exactly* how many rounds they've been using those attacks.
While in large groups people can be difficult, often individuals can be quite reasonable, especially when it is in their own self-interest.