Full charge self-consistency with ABINIT

[GingrasO]

This PR allows for charge self-consistent calculations with ABINIT. With ABINIT, one perform the Wannier90 calculation and print the DFT occupations in a file "w90_seed.abinit". Those are read by DFTTools. After performing the DMFT loop, the change in the occupations in the band basis are written in the file "w90_seed.delta_N", which is read by ABINIT.

I also corrected a few typos and changed slightly the outputted words. Feel free to remove that.

[the-hampel]

Hey @GingrasO ,

thanks for this addition. This looks very nice. I am happy to merge this as is. However, we should also start then writing some documentation that reflects and add tutorials to use the CSC interface with Abinit? What do you think.
Does this work with the vanilla version of Abinit or some upcoming release?

[GingrasO]

Hey @the-hampel,

One goal of Azin's internship is to complement the existing benchmark calculations, so we plan on adding to the tutorials as well. At the moment, this is not available in the vanilla version of ABINIT and will be pushed once a few more tests are performed. Do you think we should keep this PR as a separate branch for now? It doesn't conflict with much, so it is possible.

[the-hampel]

Hi @GingrasO ,
as you like we can keep it here as PR and merge it once ready. Or we can also merge it now and you can make changes later on. For now it will only live in the unstable branch anyway. You choose. Just let me when to merge or review again.

[GingrasO]

Hi @the-hampel,

Let's wait! Thanks.