Allow passing null for customElementRegistry

[annevk]

Apparently people are passing a ShadowRoot instance as the argument to attachShadow().

Context: https://bugs.webkit.org/show_bug.cgi?id=295174. We'll add tests as part of fixing this in WebKit.

• At least two implementers are interested (and none opposed):
  • WebKit
  • Chromium
  • Gecko
• Tests are written and can be reviewed and commented upon at:
  • See above.
• Implementation bugs are filed:
  • Chromium: Doesn't have an implementation yet.
  • Gecko: Doesn't have an implementation yet.
  • WebKit: https://bugs.webkit.org/show_bug.cgi?id=295174
  • Deno (only for aborting and events): …
  • Node.js (only for aborting and events): …
• MDN issue is filed: N/A
• The top of this comment includes a clear commit message to use.

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

---

Preview | Diff

[domenic]

Maybe add a comment here explaining why we're departing from our usual preference to only have one way of signaling absence.

[smaug----]

Yeah, some comment would good, since this is a rather odd case. Wouldn't have guessed anyone using attachShadow that way, but don't see reason why not.