Plotly doesn't recognise modern RGB syntax (+ other unexpected behaviour) #4824

Plotly does not support modern rgb(R G B / A) CSS syntax. It seems to prefer the legacy versions of rgb(R, G, B) or rgba(R, G, B, A), but it still does not support these completely (as can be seen in the examples bellow).

This can be experienced in a few different ways, such as:

1. The alpha/opacity value is ignored by Plotly even when using the legacy rgb(R, G, B, A) syntax.

>>> go.Figure().add_trace(go.Scatter(x=[1,2], y=[1,2], line_color="rgb(0, 0, 0, 0.1)")).show()
(missing opacity from plot)

2. The alpha/opacity value is also ignored when using the legacy notation with percentages

>>> go.Figure().add_trace(go.Scatter(x=[1,2], y=[1,2], line_color="rgba(0, 0, 0, 10%)")).show()
(missing opacity from plot)

3. Modern space-separated syntax raises an exception

>>> go.Figure().add_trace(go.Scatter(x=[1,2], y=[1,2], line_color="rgb(0 0 0)")).show()
Traceback (most recent call last):
...
ValueError: 
    Invalid value of type 'builtins.str' received for the 'color' property of scatter.line
        Received value: 'rgb(0 0 0)'
    The 'color' property is a color and may be specified as:
      - A hex string (e.g. '#ff0000')
      - An rgb/rgba string (e.g. 'rgb(255,0,0)')
      - An hsl/hsla string (e.g. 'hsl(0,100%,50%)')
      - An hsv/hsva string (e.g. 'hsv(0,100%,100%)')
      - A named CSS color:
...

4. The same can be seen when using the modern rgb syntax with an alpha-channel

>>> go.Figure().add_trace(go.Scatter(x=[1,2], y=[1,2], line_color="rgb(0 0 0 / 0.5)")).show()
Same exception as above

Throughout the docs (e.g., 1, 2, 3) Plotly seems to imply that any valid CSS colors are accepted. Maybe this was never the plan? If so, maybe we should at least document this a bit better and make it more explicit which values are accepted. Also, if this is the case, should Plotly raise an exception for examples 1. and 2., since the opacity values seem to be unexpectedly dropped in the background?

Thanks in advance, and happy to share more context where needed 🚀