1- Bug fix on atoms removed on the image boundary.
There should now be no problem in the atom removal. Let me know if there is.
2- I noticed there are some Tilt GBs for which there maybe a small gap difference between the main GB and the periodic image. As this happens in particular orientations I could not solve it quickly without damaging the rest of GBs. So I just added a little tolerance in gb_generator.CSL_Bicrystal_Atom_generator which reads as:
# tol = 0.01
self.atoms2[:, 0] = self.atoms2[:, 0] - norm(Or_2[0, :]) # - tol
If this is the case, uncomment the tol, change the number as you see fit.
3- I made some pep8 changes.