Étienne Polack

@antoine-levitt cleaned it a lot

Will simplify the tests with phonon PR to test more at once.

Mmm. Not ready. There is still a problem for other geometries of the unit cell. I will rewrite with loops, because the workflow is indeed too complicated at the moment...

Just a note to add checks for MPI in the main routines (tbd by Michael or Laurent)

@mfherbst this is a very rough first draft, but shows that what would need to be changed in the inner working of DFTK is reasonable. (For now I have bypassed...

ping @antoine-levitt for comments. Some stuff are still hacks, but maybe we should discuss how to do this cleanly. The setup is here and should allow to bypass DFTK's spins....

(Comments can be done quickly tomorrow with @LaurentVidal95 if you have time, it is mostly related to spin stuff.)

Failing CI for documentation may be due to current memory leak that happens on lot of projects when using v1.9.3…

Antoine and ping-ponged, and we have mostly settled on this API. Happy to have your opinion @mfherbst.

@mfherbst Could you give me some feedback on this PR? It takes quite some time to keep it up to date with the master branch.