The past PR review has shown that it needs more research.
I probably won't be doing that research (and I've written most of the most recent / popular Geometry Shader implementation).
I'm working on other projects now and I still don't have a 3DS, so I can only ever work on Citra and hardware based "unit-tests" when I have a friends 3DS.
This shouldn't be a problem as Citra is a community effort - nobody should expect me to finish what I've started. However, it also makes it impossible to give any ETA or progress in % (also because we don't know what issues we'll find in future research)
I'll still try to document what kind of tests are needed and I'll still continue fixing / working on the Geometry Shader branch if nobody does it before me. However, it's not a priority for me.
PR wise, of my original implementation (which was written more than a year ago during a couple of hours / days) there have been ~ 9 / 12 commits (?) which are in the official version already. See https://github.com/JayFoxRox/citra/pull/49
= Only 3 commits left (however, it might be split / merged into smaller / larger commits, depending on future research).