Read new /geometry/shifts/solenoid and /geometry/shifts/target CCDB tables to set target and solenoid positions independently

[raffaelladevita]

To support runs in which the target is not installed at the center of the solenoid (RG-C, RG-D), new CCDB tables were created under /geometry/shifts to store the x,y,z coordinates of the various volumes.
See https://clasweb.jlab.org/cgi-bin/ccdb/versions?table=/geometry/shifts/solenoid and https://clasweb.jlab.org/cgi-bin/ccdb/versions?table=/geometry/shifts/target

Previously, /geometry/target was used to set the positions of target, solenoid, CTOF and CND and was typically set to -3 cm. Apart from being inaccurate, this was not a problem so far because the actual target position was always equal to the solenoid center. This is not the case for RG-C and RG-D and, since the target position is used by CVT for eloss and selection of track candidates that by default is set to |z_seed-ztarget|<targetlength/2+1cm, using a wrong position leads to inaccurate eloss corrections and, more importantly, loss of good tracks.

In the new implementation:

• /geometry/shits/target is used by:
  • CVT for eloss and candidates selection,
  • FD tracking to define trajectory planes;
• /geometry/shifts/solenoid is used by:
  • Magfield to set the shift of the solenoid map,
  • CTOF and CND to set the detector position,
  • EB as a reference point for the RF correction,
  • FT to set the particle vertices.

The last 2, RF and FT, are temporary solutions to be consistent with the RG-C Summer 22 calibrations. These will be changes as follows in the future:

• EB will not use a reference vz anymore since that can be simply reabsorbed in the RF calibrations;
• FT will use the target center.

Closes #211