Chrome 119-121 had partial relative oklab()/oklch() support

[marcustyphoon]

Summary

As noted in the linked issue, this mostly copies the partial support notes for relative rgb(), hsl(), and hwb() in certain Chrome versions to document a very similar incorrect channel serialization range in relative oklab()/oklch().

Test results and supporting details

Code used and per-browser-version results are commented in the linked issue: #26041 (comment). I observed the erroneous behavior in Chromium 121, but not 122.

I'm not familiar enough with the Chromium bug tracker to find if there's a bug that should be linked here. (I wouldn't be surprised if it was fixed with the relative rgb() bug, which links to https://crbug.com/41490327.) Found using bisect-builds.py; will update link.

Related issues

Resolves #26041.

[marcustyphoon]

Chromium bisect tool says:

You are probably looking for a change made after 1223701 (known bad), but no later than 1223771 (first known good).
CHANGELOG URL:
https://chromium.googlesource.com/chromium/src/+log/36293863bf9bbc8131797eb8f4d2bafe8af3de79..96dc037087048acab5971a542fd432fae6b19855

[marcustyphoon]

Ah, shoot, that explains the weird discrepancy in initial versions. Hold on.

[github-actions]

Tip: Review these changes grouped by change (recommended for most PRs), or grouped by feature (for large PRs).