add ray multiprocessing support

[SimonBlanke]

The popular python package ray has a multiprocessing feature that could be used to run optimization-processes in parallel:

from ray.util.multiprocessing import Pool

def f(index):
    return index

pool = Pool()
for result in pool.map(f, range(100)):
    print(result)

This Pool-API is very similar to the regular multiprocessing. I will look further into if it's possible to integrate Ray so that its features can be used in Hyperactive.