[cctbxbb] new parallel_map implementation

Hi, I have switched over libtbx.easy_mp.parallel_map to use new code that is not susceptible to deadlock when large amounts of data are returned from a worker process. This also means that parameters added to parallel_map to prevent exactly this are no longer necessary (but will be respected), although unlikely to influence performance so much to warrant changing existing code. I have run all tests, but unsure how well this gets tested by regression test, so please let me know if you come across something that is amiss. BW, Gabor