Also looks like they're still planning on using existing review tools:
> We will continue to use Bugzilla, moz-phab, Phabricator, and Lando... Although we'll be hosting the repository on GitHub, our contribution workflow will remain unchanged and we will not be accepting Pull Requests at this time
Hosting it elsewhere means people are less likely to find it in the first place.
Using a different VCS means people are less likely to pick it up. Instead of starting on their contribution they have to install and learn new tooling that is irrelevant to the project itself.
Emacs is a pretty good example of this. There are plenty of people who would be interested in making improvements to the project but a much smaller proportion are willing to engage with the email-based process that's required to do so.
The crux of the issue is that the social element of an approach is just as important as the technical merit. A technically superior solution will quickly languish if no one can bear to work on it.
But Firefox is a well-known software and, at the same time, a very complex one. I don't think that somebody who struggles figuring out how to contribute via, e.g., Gitlab, will be able to contribute anything useful to the project.
People only have so much time and even insignificant barriers reduce participation.
The Emacs example is a bit more involved but still demonstrates this. There's nothing particularly difficult about submitting patches by email but the thought of doing so does put people off. The same goes for the copyright assignment, not difficult at all but reduces participation.