They're no longer used.

Adds an interface that uses fmt under the hood, which is much more
flexible than printf, particularly for localization purposes, given fmt
supports positional formatters in a cross-platform manner out of the box
with no configuration necessary.

Just so that both xgettext invocations start off the
same way, for ease of comparison.

I also removed -d because -o makes it redundant.

https://bugs.dolphin-emu.org/issues/11507

This is needed because 84cbd51 and 5efb717 added non-ASCII characters.

When building DolphinWX using cmake, msgmerge will run on the
.po files, updating the formatting of them if they aren't
already in the exact format that msgmerge uses. This means that
running msgmerge on all the .po files is necessary before
committing any updates to the .po files, because otherwise
unstaged changes are going to appear whenever anyone tries to
build that commit using cmake. The translation syncing scripts
that were used by delroth and Parlane did so by invoking `cmake`
and then `make translations`, but that isn't a good solution
for people who sync translations on Windows (i.e. me).

This commit uncomments some existing code for running msgmerge
on all the .po files, and places that code in a new script. Now,
translation syncing scripts just have to call that script before
committing, instead of using msgmerge directly or using cmake.

I'm not sure why the script was called gettextize to begin with.
gettextize is the name of a gettext executable that doesn't seem
to be related to what our gettextize script did.