This caused an issue with nested imports on Qt < 5.9

Removes all of the from __future__ imports that open and close the
brackets on the same line. This is most of them.

Runs clang-tidy pass by ref across the codebase to fix-up various
warnings we are seeing in cppcheck

Also updated class_maker.py

This commit replaces all header guards in qt/Widgets with #pragma once.

Fixes the progress arrow which previously only pointed at the first
line, irrespective of selection. Additionally removes a method which is
completely unused in the code-base, but looks like it would do a similar
thing as it's misleading.

A struct is used in case we need to add additional fields, and for the
extra readability when we dereference the map iterator. This avoids a
iter->value.at<0> in favour of having named fields of a struct.

Bumps the version of GTest to 1.10.0, removes the previous patching
steps, which are no longer needed. Removes the extra handling which is
provided in upstreams CMake, such as managing include directories

in a derived class

re #28185

This implementation is based on the implementation in MantidPlot.
A C++-based executor has been added that can execute Python code
with the option of installing a trace function to update
a progress marker on a given editor window.

The old implementation attempted to chunk up the script into
the smallest executable blocks and mark each of these as when
they were executed. There were 2 issues:
  - the chunking of the scripts was error prone as it actually
    requires a good knowledge of how Python internally handles
    indentation etc. It was fairly easy to break and future
    internal updates to Python would have required unknown changes.
  - the progress reporting resolution was far poorer than MantidPlot
    as the actual line being executed was not reported but inferred
    by what block was being executed giving a false sense of progress
    through a script.

When assigning to a list of outputs when calling the algorithm in
python, the outputs must be specified in the order they are declared,
even if they are optional outputs like IvsLam. This means that if we
want to specify the transmission workspace output we must first specify
IvsLam even if debug is false. Some of the algorithms always output
IvsLam if the name was given whereas others did not. They now always
output it if the name is given.

Note that this means you may now get an error if you don't supply enough
arguments whereas previously it was ok. It is ok if you just specify the
IvsQ output(s) but beyond that we do not know whether the output should
be IvsLam or the transmission, so we require all of them, hence the
error. This is better than it assigning an unexpected output so I think
is ok, and is the only way I can think to do this.

Re #27633

Changes related to this PR have caused test failures in the
data processor UI because they use one of the changed
reflectometry algorithms. Reflectometry no longer uses the
data process, so this is somewhat obsolete. In order to maintain
working tests I've added an option to specify the version of the
algorithm to create in the ProcessingAlgorithm so that we can maintain
the previous behaviour and previous tests.

Re #27633

This allows constraints to be empty so that if one is removed or if only one constraints is required it handles it correctly

Also dont hardcode the title string of the notficiation

This is an extra feature we want in future code but not old code. We assume
Qt4 is now just MantidPlot. It also seems to crash on macOS with Qt4

When a documentation page does not exist for a function it will now display the function index page rather than page not found.

-Change behaviour of double click so error does not appear if clicking on the category
-SelectFunctionDialog changed to be modal

If the user enters an incorrect function into the search bar it will now give an within the dialog and keep the dialog open

When getting the function name if a nested group name is selected instead of a function it will return an empty string (the same as it does for top level groups)

Fitting now conforms to user fit range
TF asymmetry fits work in single fit mode
The list of fit workspaces now preserves order

The fitting tab was keeping a reference to two function browsers, which had to kept up-to-date with each-other. This issue has been addressed by introducing a single function browser.

From the workspaces toolbox in plot and workbench