I don't see how this would actually help improve the parts of GNOME that need improvement.
If the Gnome devs see that a particular extension is installed often (i.e. GsConnect, a sound output chooser that's actually useful, etc.) then they might build that feature into the system directly so users don't need extensions any more.
I'm sure they can collect more information if they want to but I don't know if that information would actually be useful to the Gnome team at this point. I'm guessing they're looking for very specific things to optimize for (i.e. Flatpak support, hardware support, extensions everyone uses to compensate for missing features).
I just think they could have gone way broader with the set of reported data, and the people willing to help wouldn't mind.