Bug fix contributed by @dehond (Issue #126 , PR #127) :
- when calculating total polarisability, sign of podermotive (dynamic) polarisability was wrong (38ff910)
- docstring explaining order of returned polarisabilities had permuted places of core and podermotive poarisability (882a60f)
