dials.reciprocal_lattice_viewer: correctly record beam centre panel for multi-panel detectors

[dagewa]

When models are loaded, the beam centre is calculated, and if this is succesful, the result is written back to the beam centre controls in the GUI. Unfortunately, the panel number was not written back to its control, which defaulted to 0. This meant that whenever the relp positions were next calculated, it would do so under the assumption of the beam centre being in panel 0. This is the cause of the errors reported with data from I23 (#2571).

This PR correctly updates the control when models are loaded, so that the geometry is correct when next recalculated.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (2e18ef5) 78.52% compared to head (14b4e59) 78.53%.
Report is 2 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2572   +/-   ##
=======================================
  Coverage   78.52%   78.53%           
=======================================
  Files         609      609           
  Lines       75034    75034           
  Branches    10706    10706           
=======================================
+ Hits        58923    58925    +2     
  Misses      13943    13943           
+ Partials     2168     2166    -2     

[dagewa]

Windows test failures are something unrelated. Merging this now to make it available for beamline scientists.