Need for this highlighted by performance test failure (GetDetOffsetsMultiPeaksTestPerformance).

SANSRunWindow::loadCSVFile : Take the first line of the csv file and
insert it in SingleMode MWRunFiles.

re #6225

re #5942

* SANSRunWindow.ui: Add widgets inside Geometry Tab that allows to show/edit
  the centre for the high/low angle bank.
* SANSRunWindow.cpp: Interact with the changes in SANSRunWindow and also
  with the changes in the script in order to support both detectors.
* isis_reducer.py: The changes made here is to support the definition of the
  centre for both detectors, allowing to get its position get_beam_center
  and to define their positions. It does extend the class Sample to allow
  reloading the data to move the detector bank component to the correct
  position in order to apply the reduction.
* isis_reduction_steps: Extend parsing the USER FILE, in order to accept
  the commands: SET CENTRE/[MAIN/HAB].
* ISISCommandInterface.py: Adapt the FindBeamCentre method to support both
  detectors, adapt the WavRangeReduction to reload the dataset for
  moving the hab detector to its center in order to correct use this
  parameter for the reduction. Extends the method SetCentre.

Release the memory allocated for listing the spectra related to the detectors IDs.
RE #6157

The prolem actually occured with the new reader. When the PV framework was
deciding which reader to user (running canReadFile) it passed over the
vtkNexusPeaksReader. The canReadFile implementation here opened the nexus
file, but did not close it. Thefore it could not be reopened when the
correct reader was finally chosen by ParaView.

in an analogous way that it is done with cxxtest_add_test.

The Python test directory is trashed and completely regenerated to
guarantee it is up to date.

Test has moved to system tests.

Collapsed many functions to two.

This doesn't actually fix the version it just makes it easier to do
in the future.