bug#66187: read-file-name unexpected behavior when MUSTMATCH is a function

[Drew Adams]

> The use of a function as `require-match` is
> brand new in Emacs-29, so I think it's not
> too late to fix it.  I think rather than
> fixing the doc we should fix the behavior 

If it's not too late to fix what was introduced
in Emacs 29, then how about not breaking the
behavior of REQUIRE-MATCH at all, and instead
just add a separate argument for what a function
value of REQUIRE-MATCH does in 29.1?

(As opposed to what a function value of
REQUIRE-MATCH has always done before Emacs 29,
i.e., act like any other non-nil, non-`confirm*',
non-`t' value.)