GN: finished debugging the hodoscope start time calculation. Results match 1-to-1 the corresponding "engine" result.

Started with Simon's code, but reworked:
	Use podd style arrays in RVarDef.
	Get plane names (1pr, 2ta, 3ta, 4ta) from parameter file.
	Get block counts from parameters instead of hardwiring.
	Rename histogram names to be A,B,C,D 1-13, instead of 1-52

  It has not been check that we are getting a good start time yet.
  Drift time and distance added to tree
  Change DC plane names from 1, 2, 3, ... to 1x1, 1y1, ...
    This is so that the parameters holding the time to distance maps
       can be found.  (The parameters are e.g. hwc1x1fract)
    Need to find a way not to have to hard code these plane names.
       Either use wire angles (alpha) or some kind of parameter name mapping
  Changed output.def to match new plane names

Change shower plane names from 1z, 2z, 3z, 4z to 1pr, 2ta, 3ta, 4ta
  to match the names used in the parameters.
Fix up histogram names for new plane names.
Get rid of duplicate THaNonTrackingDetector::Init call.

pedestal event analysis at Decode, no cut needed on histograms need to assure
that events with HMS (evtype 1 and 3) get histogrammed.

don't add to histograms.

Put map files in MAPS subdir.
Define correct map for run 50017 in test.database.
Add a script to display drift chamber hitmaps.
Fix make_cratemap.pl so that it makes an entry in db_cratemap.dat
  for all Fastbus slots.

be made either in output.def or by analyzing root tree.