Switch to a more accurate color vision deficiency simulation model

[nburrus]

Hi! Recently I've been reviewing the CVD simulation models used in opensource software, and it turns out that the ColorMatrix values from colorjack.com are very inaccurate (I saw that these are the values used in color-blind-simulator).

I wrote a detailed article about that, but in short this is what the author later said about these matrices (links from MaPePeR, the code used by Coblis):

You're right, the ColorMatrix version is very simplified, and not accurate. I created that color matrix one night (http://www.colorjack.com/labs/colormatrix/) and since then it's shown up many places... I should probably take that page down before it spreads more! Anyways, it gives you an idea of what it might look like, but for the real thing...

So I've started a kind of crusade to help stop the spread :)

The good news is that there are pretty good alternatives that also just use a single 3x3 matrix and thus can be implemented with SVG filters. I've actually created a PR in another project that uses SVG filters Colorblindly that does just that. It updates the matrices to use the proven model of Viénot, Brettel & Mollon Digital video colourmaps for checking the legibility of displays by dichromats. And also importantly it adds a color-interpolation-filters="linearRGB" attribute to the filter so that the matrices get applied in the linearRGB space and not sRGB.

Hope this can help!