Remove width as relevant mutation for image

[TimvdLippe]

An engine that implements this will start to cause
several tests to either flake or incorrectly pass: servo/servo@2e584db

Fixes #9371

• At least two implementers are interested (and none opposed):
  • Chromium
  • …
• Tests are written and can be reviewed and commented upon at:
  • https://wpt.fyi/results/html/semantics/embedded-content/the-img-element/relevant-mutations-lazy.html?label=experimental&label=master&aligned
  • https://registry.khronos.org/webgl/sdk/tests/conformance/textures/misc/texture-upload-size.html?webglVersion=2&quiet=0&quick=1
• Implementation bugs are filed:
  • Gecko: …
  • WebKit: …
• The top of this comment includes a clear commit message to use.

(See WHATWG Working Mode: Changes for more details.)

---

/images.html ( diff )

[TimvdLippe]

@zcorpan Since you opened #9371 would Gecko be supportive of changing the spec here? When we implement it as-is, then Servo would have significant flakiness in two test suites.

[annevk]

If the main engines already do this, this doesn’t need explicit support. Might be good to mention that in the final commit message.

[TimvdLippe]

Well, I am not confident to claim that. When I looked up the relevant tests, Gecko and Webkit might actually implement this: https://wpt.fyi/results/html/semantics/embedded-content/the-img-element/relevant-mutations-lazy.html?label=experimental&label=master&aligned