[AMD2 WD: COLR v1] Address "TBDs" in working draft regarding open issues on radial gradients

[PeterConstable]

The working draft presented at the January SC29 meeting (m55831-COLR_v1_updates-14496-22-AMD2.docx) had "TBDs" calling out two open issues regarding specification and behaviour of radial gradients:

TBD: What should the expected behavior be in the case of both circles exactly overlapping with r > 0? (Only the extensions get painted, but how?) _TBD: We still need to specify required behavior for the case in which the transform really flattens the two circles, and the centers, to a line.>

These need to be addressed in the next working draft.

[PeterConstable]

The draft in the Google Fonts repo has been updated to address this issue. The following is the main pull request for that change:

• For radial gradient, clarify if circles are identical
• Clarify result if radial gradient circles are flattened

While working on this, it was noticed that visual Moiré patterns can occur if the circles specifying a radial gradient are not quite identical. Also, in a variable font, if the circles are varying, then a sharp transition can occur. It was thought to be a good idea to call these things out in the text. The following PR has changes for this. (Addition of figures required renumbering of all subsequent figures.)

• add note for rad grad re sharp transitions if rel'nship of circles vary

Also, another small wording change in the radial gradient algorithm was made:

• Revise wording for radial gradient algorithm

[vlevantovsky]

Proposed changes have been incorporated as part of the input contribution m56341.