Unify MDT data access for L2 and EF muon algorithms in athenaMT trigger
These changes mean we use the offline algorithm for the MDT data decoding in the athenaMT trigger chains. This avoids running the decoding in both the L2 and EF steps. There are also a couple of bug fixes, including putting the EF data preparation into RoI mode, which changes the trigger output. I tested by running the muon menu ctest and the legacy trigger ART test. Both run fine.
Edited by Mark Andrew Owen