Fix for C++ conversion error in plotting

[jellybean2004]

Description

Fixes #3740

The issue occurred because None was passed as a parameter to plotRequestedSignal.
In src/sas/qtgui/Utilities/GuiUtils.py:106, the signal is defined to accept list and int. Since None is not an int, it triggered a C++ type conversion error when the signal was emitted, as Qt is implemented in C++.

PySide does not support Python-style union type hints (e.g. int | None) in signal definitions, so supporting both types directly is a rather cumbersome implementation.
As a workaround, the fix passes a valid integer value instead of None:

• 0 for perspectives that do not support tabs,

• self.tab_id for those that do.

Although a deeper look showed that tab_id is not actually used in the connected slot, it has been left intact to preserve the existing signal interface.

How Has This Been Tested?

Checked by plotting in all the affected perspectives (SizeDistribution, Inversion, Invariant). The error does not appear anymore.

Review Checklist:

[if using the editor, use [x] in place of [ ] to check a box]

Documentation (check at least one)

• There is nothing that needs documenting
• Documentation changes are in this PR
• There is an issue open for the documentation (link?)

Installers

• There is a chance this will affect the installers, if so
  • Windows installer (GH artifact) has been tested (installed and worked)
  • MacOSX installer (GH artifact) has been tested (installed and worked)
  • Wheels installer (GH artifact) has been tested (installed and worked)

Licensing (untick if necessary)

• The introduced changes comply with SasView license (BSD 3-Clause)

[rozyczko]

I'm slightly confused how this could be the solution.
plotRequestedSignal's only slot is in GuiManager

self.communicate.plotRequestedSignal.connect(self.showPlot)

showPlot passes the second argument (id) to displayData, not doing anything to it.

self.filesWidget.displayData(plot, id)

Sunsequently, displayData, although accepting the argument

def displayData(self, data_list, id=None):

does absolutely nothing with it. The whole id passing chain is completely redundant at this point (it used to be a few rewrites/fixes earlier though!)

[rozyczko]

Never mind. I didn't read your description properly :) This makes sense in the context of proper type checks.
However, you might want to consider just removing this argument throughout, so no conversion is taking place. Nor confusion

[jellybean2004]

Since the tab_id parameter is obsolete, I have removed it from the definition and all the calls.

[DrPaulSharp]

All looks good, rebase and merge when ready.

[rozyczko]

Looks good now!

[wpotrzebowski]

Tested on Mac from installer. It works as expected