Sort members of a WorkspaceGroup by workspace name

 Vesuvio - VesuvioThickness algorithm

Refactor MatrixWorkspaceTest::createImage(...).

Refs #15816

Refs #15816

Refs #15816

Refs #15816

Refs #15816

Refs #15816

Refs #15816

Merge pull request #15949 from mantidproject/15936_ProcessingInstructions_CreateTransmissionWorkspaceAuto

CreateTransmissionWorkspaceAuto: fix default ProcessingInstructions

Fix error initializing python with broken files

Reduce the size of GluGeometryHandler

Correct with sqrt

Fit plot improvements in MuonAnalysis: keep fit curves, set current data

re #16017

If the python initialization came across a file which could not be
read (e.g. emacs intermediate save file), all of python would fail
to initialize. This will only affect people that develop code in a
directory that mantid checks for algorithms.

Add an empty/none marker option in pyplot.plot()

Merge pull request #16010 from mantidproject/15950_documentation_for_pearl_powder_diffraction_script

Documentation for Pearl powder diffraction script

Refs #15965 LoadBBY and PatchBBY now support monochromatic neutrons

Fix bug with complement rule parsing in Object::procString

System test for Pearl powder diffraction script

re #16017

Default behaviour for the PeakPickerTool is to clear old fits off the
graph when a new fit completes. Muon scientists would prefer to
keep the old fits on the graph and remove them manually when it gets
cluttered ("clear fit curves").

re #16017

This caused the bug in the issue when a graph had:
- curve 0: fit to dataset A (old)
- curve 1: data from dataset B (new)
and the PP tool initialised from dataset A, meaning that the workspace
A was the one shown in the "Data" tab.

(Rationale:
It doesn't make sense to initialise the PP tool from a fit.
It's a tool for fitting data, and why would you fit a fit curve?)

re #16017