Coulomb bare -1/r potential for H. For testing purposes, or for all cases
where there is no real advantage in using a PP for H (e.g. if there are other
atoms with "hard" PP). Should be reasonably converged around 70 Ry or so.
Note that this PP is internally tagged as "PBE", even if it is actually
independent on the specific XC functional used. Use input option "input_dft"
if the code complains about "inconsistent dft".