Fix help button on Algorithm dialog not scaling

update group calibration script

Fix muon GUI phasequad bug

Renamed the Muon Results Table Tab to Results Tab

Fix unstable PythonInterfacesTest on macOS

Change ConvertToReflectometryQ to work with detectors at twoTheta

Expose workspace readLock to python to prevent segfaults

This is because it was capped to 25px rather than letting Qt figure it
out.
Re #32324

Main Cleanup of Muon Analysis release notes

Also hads a call to process internal events such as deletions
so that they are not delayed until too late in the program
lifetime where they can cause a segfault

Github pull request 30242 introduces delayed refresh. This is no longer
necessary with the addition of non-blocking QAppThreadCalls. See gitub
pull request 32204 for more details.

Added final set of screenshots, updated text for continuity and added additional links where appropriate.

synchronized ADS event handling in WorkspaceSelector

Added screenshots, reorganised as some content under wrong headings, missing content added, spellcheck.

Guarantees a sequential flow

Co-authored-by: Pete Peterson <peterfpeterson@gmail.com>

Absorption and Multiple Scattering concept page

Fix flaky NOMAD test

Manually prep the filenames in a temp directory, instead of relying on
generated tmp filenames. This is because tempfile will also hold a
handle to the referenced file too which prevents us writing to it in the
Alg.

This eliminates any racey-ness introduced by using the ADS, as we can
just use these as pointers directly.

Windows was failing with an OS error if the file was locked.
Fixes possible causes for a flakey NOMAD test pre-emptively:

- Use context manager as NamedTempFile will do a better job cleaning up
than us
- Clear the ADS after we finish
- Use pointers to workspaces instead of hard-coded names
- Use validate in the system test instead of assertTrue, as per other
system tests

Re #20327

Add a version number to the underlying classes to switch between the old
and new behaviours

Replace the original test and data and move the new functionality to a
v2 test. Note that there was already a v2 of the data so the new
reference files are v3 even though they relate to v2 of the algorithm.

Note that the new calculation is tested in the ReflectometryTransform
tests so does not need to be tested in v2 of ConvertToReflectometryQ2.

Re #20327

In theory, I would expect to be able to produce the same results as
previously, so this is not ideal. But to do that would require moving
the instrument very precisely to the correct position and this would
probably require a calibration file (which we don't currently have an
easy way of applying in Mantid). This is beyond the scope of this PR
because this is a very old system test which will probably be superseded
in the future anyway.

For now, I've therefore made an approximate change
to the instrument position, which makes the output look roughly
sensible, but the counts are different, so I have had to update the
reference files as well. This should be fine for now for the purposes of
testing just to ensure we catch any behaviour changes, even though this
example is unfortunately not such a good example scientifically any more.

Re #20327