Fix specifying detectors to mask in SANS detector banks

[Steve Williams]

Get "Default Mask" of the SANS GUI "Masking" tab working correctly. Fix the syntax for masking in the user file as below.

From: Heenan, Richard (STFC,RAL,ISIS) 
Subject: RE: SANS interpolating monitor spectra ETC

Steve,

Will go with having

MASK/FRONT  for any row & column, and time masks for the front detector

MASK or MASK/REAR for rear detector.

DET/FRONT or DET/REAR just control the drop down in the gui as to which detector we are about to reduce the data for.

One day, not that soon, we might like  DET/BOTH and DET/MERGE options on this list to do both detectors into separate workspaces; and both into separate and a composite workspace.

[Steve Williams]

before change test mask file

[Steve Williams]

There are apparent glitches in the "Display mask" after I've loaded the attached user file.

I'll leave implementation of the DET/BOTH, MERGE to after the refactoring ticket #1686

Time estimate: To read the new user file syntax and change _applyMasking() and ViewCurrentMask(), 1 hour. Test that it definitely all works, SANS2D, LOQ, front back, masking rows columns, shapes and spectra 3 hours. To copy the changes over to the new refactored version and check, 1 hour. I don't have an estimate on how long it will take to fix view mask. It is possible that the above work will already fix that. Total 5 hours.

[Martyn Gigg]

13. Gigg time estimate: I probably wouldn't put as long on the testing estimate. 2 hours for that seems plenty so that puts my estimate at around 4 hours.

[Steve Williams]

(In [6874]) Fix a SANS masking bug -the first non-detector spectrum in the empty instrument is different from in a normal run re #1750

[Steve Williams]

(In [6877]) Fix tests re #1750

[Stuart Campbell]

This ticket has been transferred to github issue ​2597