> But yes, not everyone can or will write good commits.
some people treat commits as meaningful units of independent review, and some people treat them as savepoints and the PR as the only meaningful unit of review, it's a distinction of process, not purity -- both approaches are totally fine, one is not better than the other
git and commits and prs are means, not ends
That's exactly it. What happens in a private branch is an implementation detail and reflects personal work style. Policing that is counterproductive.
No one said anything about "policing" anything. I'm not telling anyone how to write PRs, I'm just suggesting that if we had better tools we'd get "better" PRs for values of "what happens in this 'private' branch is more than an implementation 'detail' but a useful story and a useful documentation of the process". You don't need to agree that is "objectively" or "universally" a "better" way to make PRs for everyone and every project, but I'd hope you could at least respect that it's a nice goal that some of us have at least some of the time and why we would like PR tools that respect that approach as much as they seem to already respect your "no one cares how the sausage is made" approach.