Modify charge dialogs qml

[RobBuchananCompPhys]

This PR aims to implement the three dialogs "Scale Charges", "Smooth Charges" and "Reduce Sig Figs in Charges", in QML.
I have used a generic ModifyChargesModel to handle the resources corresponding to each operation, since all dialogs share a common, simple format of components, and similarly updated the dialog class to handle connections between all three QMLs and the model.
Therefore, this supersedes (and incorporates) the work I did recently on the ScaleChargesDialog/Model.

[trisyoungs]

Something corrupt / missing in the qml files, or at least their location as I get this error (as does the CI):
ninja: error: '/home/tris/src/dissolve/src/gui/qml/ModifyChargesDialog.qml', needed by 'src/gui/qrc_main.cpp', missing and no known rule to make it

[trisyoungs]

@RobBuchananCompPhys An update from my machine. Upgrading my distribution has now furnished me with Qt 6.4.2 which solves all of the odd errors with text / widget colouring I was seeing. I also now understand the font scaling issue - I still see odd results when setting font.pixelSize but font.pointSize gives the expected results when set to 11. This must be down to a mis-calculation of screen DPI or something similar at the Qt end since my display is not scaled, but it seems that the recommended approach is to use pointSize wherever possible.

[trisyoungs]

@RobBuchananCompPhys An update from my machine. Upgrading my distribution has now furnished me with Qt 6.4.2 which solves all of the odd errors with text / widget colouring I was seeing. I also now understand the font scaling issue - I still see odd results when setting font.pixelSize but font.pointSize gives the expected results when set to 11. This must be down to a mis-calculation of screen DPI or something similar at the Qt end since my display is not scaled, but it seems that the recommended approach is to use pointSize wherever possible.

We can set this in the base D.Text class, presumably, in order to provide a consistent result?

[trisyoungs]

Looking good. Like the unit test!

I still have a question over whether we really need to stash a pointer to a ModifyChargesModel anywhere - rather, can't we just new one when we need it (see my comment)?

[trisyoungs]

Looking very nice now. Regarding the font.pointSize issue, do we then just remove that for now until we understand the (cross-platform) effects a bit better?

Also, I have pushed a small commit on to the PR which just moves the declaration of the model in gui.h to more of its own section. Felt silly explaining that and requesting a change!

[RobBuchananCompPhys]

Looking very nice now. Regarding the font.pointSize issue, do we then just remove that for now until we understand the (cross-platform) effects a bit better?

Also, I have pushed a small commit on to the PR which just moves the declaration of the model in gui.h to more of its own section. Felt silly explaining that and requesting a change!

Thanks @trisyoungs !
I have asked that @rprospero takes a look at how this looks post-font.pointSize: 11 commit just to get another opinion, currently sounds like it may well be a graphics issue. But I'd certainly consider holding out and reverting that particular commit if we can't pin down exactly why the odd size changes are occurring on diff platforms.

On another note, I noticed that the simulation menu also utilises Reduce significant figures in charges... function, so I'll add a new commit which essentially does the same thing in the species menu for sig figs. We can maybe discuss that one more in stand up tomorrow.

[rprospero]

@trisyoungs @RobBuchananCompPhys Here's the dialog on my machine.

[RobBuchananCompPhys]

@trisyoungs @RobBuchananCompPhys Here's the dialog on my machine.

Looks promising! Shall we merge?