"Empty case ranges such as case -1 ... 0 are valid, but recommended to be diagnosed. "

Why is that an empty range?

Because I'm an idiot and meant to write 0 ... -1.

Re the "A noteworthy quirk of the C2y feature" paragraph, that was documented decades before it got into C2y:
https://gcc.gnu.org/onlinedocs/gcc/Case-Ranges.html
It might be worth linking to that page for the implementation experience talking about the extension.

Thanks, added a link, though not in the implementation experience section.