This is true.
> and this is what makes it possible to find bugs this way
No, this is not true. In fact, it is the opposite of true.
If a program cannot rely on behavior, it cannot rely on having the behavior that causes the bug to be visible.
For this reason, signed overflow is essentially a solved problem, while unsigned wraparound will be the source for many interesting bugs - exactly because it is not UB!
Does the compiler faithfully compile the source code with sanitizers? Of course not, so we still cannot rely on them.
> So you can not distinguish between intended wraparound and incorrect wraparound.
This is exactly why I implemented both types in my arithmetic. The one I used is the one I intended.