Allow partial enums

[alvestrand]

I was looking for whether we permit partial enums like we do extended interfaces, and found this bug:

https://www.w3.org/Bugs/Public/show_bug.cgi?id=27048

This doesn't seem to have concluded, but I couldn't find a similar bug in this repo.
FWIW, I think having partial enums is no more dangerous than partial dictionaries or partial interfaces; all of them can lead to name collisions if not well maintained. So I'm (so far) in favor of adding them.

[saschanaz]

I'm seeing some more needs:

• WebXR Augmented Reality Module: People want to add immersive-ar to XRSessionMode from WebXR Device API - Found from Validate IDLs in idl_test() web-platform-tests/wpt#18382
• Clipboard: People want to extend permission types defined by Permissions spec with new clipboard-read and clipboard-write - Found from Add Async Clipboard API permissions microsoft/TypeScript-DOM-lib-generator#735

[annevk]

Even though it's more coordination work, I still think encouraging everyone to maintain centralized lists for these is good, per the rationale in w3ctag/design-principles#99.

[marcoscaceres]

Agree with @annevk. More coordination seems preferable as it provides more checks and balances.

[mounirlamouri]

I'm seeing some more needs:

• WebXR Augmented Reality Module: People want to add immersive-ar to XRSessionMode from WebXR Device API - Found from web-platform-tests/wpt#18382

This one is fairly simple to handle given that both specifications are part of the same WG and worked on by the same people.

• Clipboard: People want to extend permission types defined by Permissions spec with new clipboard-read and clipboard-write - Found from microsoft/TSJS-lib-generator#735

The Permissions API already has a list of all permissions. Clipboard should just adds itself in there.

[saschanaz]

Filed immersive-web/webxr-ar-module#32 and w3c/clipboard-apis#101 to fix them.

[henbos]

Here's an example where we ran into this issue and added a comment, not great:
https://w3c.github.io/webrtc-extensions/#rtcicecredentialtype-enum

[saschanaz]

@henbos Were there any blockers that prevents the WebXR way? (It added the extension to the original enum definition with a prose, which explains that its behavior is defined in another spec.)