Complete PySCF twist implementation

[prckent]

As per the discussion in #5073 , a few more updates to the converters are needed for problem free support of twists and supercells.

When completed, to guard against incorrect minus signs and similar small mistakes, it is would be worth verifying that the correct HF energy is obtained at a few twists. e.g. Run single k-point at gamma, x, somewhere arbitrary, similar to the checks done for plane waves/splines.