You're always going to have trouble if you try to "enforce" semantic versioning, because it's a social commitment by project maintainers, not a theorem that can't ever possibly exist about "breaking changes" (which has a fuzzy, social definition).
The best option for projects is probably to loudly declare that you are not using semantic versioning, and then use it anyway, on the good-faith basis that is the only actually possible way to use it. Declaring that you don't use it will head off trouble from people who expect it to be an impossible magic constraint, while using it will help people make sense of your version numbers.
100% agree. Semver is a _social_ construct not a _technical_ one. There is Semver the spec as originally designed, and Semver as it is currently observed. Semver has become something other than it was originally intended, likely because the idea of "breaking changes" ends up being way more encompassing than people wanted to admit [1]. To ignore that reality is a futile fight.
The best option for projects is probably to loudly declare that you are not using semantic versioning, and then use it anyway, on the good-faith basis that is the only actually possible way to use it. Declaring that you don't use it will head off trouble from people who expect it to be an impossible magic constraint, while using it will help people make sense of your version numbers.