Introduced a handling mechanism for event range parsing errors in multithreaded
event service. The errors are reported back to the Pilot the same way as it has
been done in multi-process event services.

Modified MT event service related CLI for athena:
   * --mtes activates multithreaded event service if used in combination with --threads=N,
otherwise has no effect
   * --mtes-channel sets the name of an IPC channel between AthenaMT and Pilot. Has effect
only if used in combination with --mtes

Removed the redundant AtlasThreadedEsJob.py. All configuration tasks are now handled by
AtlasThreadedJob.py.

Finally, did a minor code cleanup in AthenaHiveEventLoopMgr