Fixes #3750 - MouseEnter/Leave

[tig]

I've decided to rebase v2_2491->Overlapped, which was here:

• Fixes #2537, #3691, #3406 - New Overlapped Design, mouse/keyboard sizing - Progress on #2491 #3705

... onto this PR given this PR is dependent on it. This will make is so there's only one big PR merging into v2_develop.

Fixes

• Partially Addresses Get rid of Toplevel - Introduce Runnable and Overlapped instead #2491 by
  • Finishes ViewArrangement.Overlapped support
  • Fixes Add mouse/keyboard resizing of views #2537
  • Fixes Slider: API for setting options is clunky #3735
  • Fixes Fix focus and keyboard nav with Adornments #3406
  • Fixes Move all rendering out of Toplevel, FrameView, Window, etc... into View. #2484
• Fixes MouseEnter/Leave is broken for Views with subviews #3750
• Implements Hover mode
• Revamps ColorSchemes and removes more legacy Color code.
• Addresses some of Refactor MouseEvent and MouseEventEventArgs to simplify #3029

Proposed Changes/Todos

• Remove temporary code in OnDrawContent and get z-order correct.
• Enhance Bars scenario to illustrate Popover menu concept
• Enhance Navigation scenario with ArrangementEditor; like AdornmentEditor makes it wasy to change view settings related to arrangement for testing/demos.
• Clone Navigation scenario to Arrangement scenario - one for each
• Implement support for ViewArrangement.Resizable
• Fix Draw
• Remove more code from Toplevel - See how far I can get in removing Toplevel allogether without re-doing all the Runnable stuff (for a later PR focused on IRunnable).
• Fix Add mouse/keyboard resizing of views #2537
• Update docfx/docs and other API docs
• Refactor FindDeepestView into GetViewsUnderMouse. -> GetViewsUnderMouse ().LastOrDefault () is equivalent to FindDeepestView
• Change MouseEnteredView to List <View?>? ViewsUnderMouse
• Add internal Application.RaiseMouseEnterLeaveEvents () called by OnMouseEvent - manages ViewsUnderMouse and raises events as appropriate.
• Update Mouse Scenario to really stress GetViewsUnderMouse and events
• Add more unit tests for GetViewsUnderMouse
• Re-add demo of wantcontinuousbutton press in Mouse scenario.
• Remove MouseEvent params from MouseEnter/Leave events; there's no need for them.
• Determine if MouseEnter should be cancellable. Yes.
• Enable `HighlightStyle.Hover" <--- @bidisp!
• Re-enable ViewDiagnostics.Hover
• Update config.json with more themes and change colors to make prettier
• Refactor Color handling to exclusively use ColorStrings instead of legacy ColorName values.

For future PRs

• Enable Runnable concept
• Fix Tiled mode: need a way to declare tiled view size dependencies #3741
• Fix Navigation Scenario: Add a TabStop Editor  #3742
• Fix Popover - A consistent way of enabling a Subview to popup outside of a View #3691
• This PR re-broke the cursor flickering issue (Long-standing repaint-flicker issue. #3408). Will fix this in Re-factor Cursor handling per ./docfx/docs/cursor.md #3444
• Create a scenario that illustrates an "MDI" like app (replacing BackgroundWorkerCollection). Show how MenuBar and StatusBar can be in any View, even multiple.
• Fix Subviews in Adornments that are not ViewArrangement.Fixed are not supported #3743
• Fix Remove need for TileView - Use LineCanvas to draw borders and auto-join borders #2483

Pull Request checklist:

• I've named my PR in the form of "Fixes #issue. Terse description."
• My code follows the style guidelines of Terminal.Gui - if you use Visual Studio, hit CTRL-K-D to automatically reformat your files before committing.
• My code follows the Terminal.Gui library design guidelines
• I ran dotnet test before commit
• I have made corresponding changes to the API documentation (using /// style comments)
• My changes generate no new warnings
• I have checked my code and corrected any poor grammar or misspellings
• I conducted basic QA to assure all features are working

[BDisp]

I think the shadow button shouldn't move to the right, like as it isn't moving to the bottom, when it's pressed. Only the shadow effect should be hidden when the button is pressed. In my opinion the button is always in the same place and only the light is placed on top/left side chaining a shadow on the bottom/right. If the light is projected in others directions the shadows will be propagate accordingly.

[BDisp]

Starting reviewing this wit some observations. If necessary I'll will add more.

[tig]

I think the shadow button shouldn't move to the right, like as it isn't moving to the bottom, when it's pressed. Only the shadow effect should be hidden when the button is pressed. In my opinion the button is always in the same place and only the light is placed on top/left side chaining a shadow on the bottom/right. If the light is projected in others directions the shadows will be propagate accordingly.

My opinion is the current effect looks great.

This was discussed here:

#2144 (comment)

And the code has this comment:

It would be trivial to add a CM property to control this, but we'd still need to have a default.

I enabled the effect by default in this PR so that it was in everone's face and we could get good debate going.

The right place to further opine on it is here:

#3493

[BDisp]

I already noted that unit tests are extremely slow in this PR.

[tig]

I already noted that unit tests are extremely slow in this PR.

Fixed.

[BDisp]

These two images bellow refer to the same inline data in the GetKeyChar_Shifted_Char_From_UnShifted_Char unit tests. They show different bitwise masks. That why it's showing as not run tests. I think the reason for this is how the bits are set because the KeyCode is a enum using the [Flags] attribute. He may represent a combination of KeyCode values but mustn't have a new variant value. I think this is very strange and that's why the Test Explorer sees it as a new unit test. I commented the others InlineData to only show the one below.

[InlineData ('9', ')', (KeyCode)')' | KeyCode.ShiftMask)]