ENCUTGWSOFT soft cutoff for Coulomb kernel

ENCUTGWSOFT= [real] (energy cutoff for response function)

Default:
ENCUTGWSOFT=ENCUTGW $\times 0.8$	for ALGO=ACFDT
ENCUTGWSOFT=ENCUTGW	else

The flag allows to truncate the Coulomb kernel slowly between the the energy specified by ENCUTGWSOFT and ENCUTGW. This usually leads to much smoother energy-volume curves in AC-FDT and MP2 calculations. The modified Coulomb kernel is in this case:

$$v_{\bf G} = \frac{4 \pi e^2} {\vert \bf G\vert^2} \frac{1}{2} \le... ...t^2 }{2 m_e} - {\tt ENCUTGWSOFT}}{ {\tt ENCUTGW - ENCUTGWSOFT}} \right) \right)$$   for$$\quad \frac{\hbar^2 \vert\bf G\vert^2 }{2 m_e} > {\tt ENCUTGWSOFT}$$