LU-2624 ptlrpc: improve stop of ptlrpcd threads

LU-2624 ptlrpc: improve stop of ptlrpcd threads

This fix improves the stop of ptlrpcd threads so that
it occurs in parallel. On machines with a large number
of cores this significantly reduce the time of unload of
ptlrpc module.

Signed-off-by: Gregoire Pichon <gregoire.pichon@bull.net>
Change-Id: I744de5d9887bfaa4c7a742d34923680e3a536ee7
Reviewed-on: http://review.whamcloud.com/5039
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Faccini Bruno <bruno.faccini@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>