Fix WebGL vertex stroke colors for retained mode + some shape modes

[davepagurek]

Resolves #5926

Changes:

• Makes sure vertex stroke colors are duplicated along with fill colors when using QUADS, QUAD_STRIP, etc
• Uses separate buffers arrays representing the data sent to the line shader (lineVertexColors) from the per-vertex color info (now called vertexStrokeColors)
  • This is because we must ALWAYS supply vertex color attribute data to the line shader, otherwise the buffer's stride+offset will be wrong
  • We now determine whether to use per-vertex stroke colors by checking if the user-supplied per vertex colors (vertexStrokeColors) is empty, now that the one used for rendering will always be filled
• Updated the TESS mode per-vertex interpolation code to use less hardcoded lengths. We don't actually need to change the length here since TESS only applies to fills, not strokes, but I've kept this minor refactor in anyway to make the code more maintainable for if we need to add more in the future, as I initially thought we would here

Screenshots of the change:

Before	After

Live: https://editor.p5js.org/davepagurek/sketches/DIb8nX4-C

PR Checklist

• npm run lint passes
• [Inline documentation] is included / updated
• [Unit tests] are included / updated

[inaridarkfox4231]

I see, it was better to separate the array used to configure the vertex colors and the array sent to the shader... Thanks for the fix!

[inaridarkfox4231]

I'm sorry, I'm using "lineVertexColors" to get the vertex colors of the lines in bezierVertex() and quadraticVertex() which was recently merged, so I feel like I have to change that to "vertexStrokeColors" as well.

[davepagurek]

ooh good catch, I'll change it there too!

[inaridarkfox4231]

And...for bezierVertex() and quadraticVertex(), I didn't add a unit test for line vertex color interpolation, but that might have prevented it...wouldn't it be nice to add that as well?

[davepagurek]

agreed, I'll add some new tests when I make those changes.

[inaridarkfox4231]

Thank you for even preparing a test for _useLineColor! It's certainly safer...