VirtualKeyboard spec feedback

[yoavweiss]

Looking at https://w3c.github.io/editing/docs/virtualkeyboard/, there are a few points that can be made clearer:

• In section 3, "If the browsing context's active window" is inaccurate. It'd be better IMO to replace it with "this's relevant global" while asserting that it's indeed a Window (which it should be, as the method is only exposed on a Window).
• I'm assuming that step 5 in both show() and hide() should be run in parallel, and the event doesn't fire synchronously. You should make that explicit. Depending if step 4 takes time, you may also want to run it in parallel as well.
• overlaysContent is missing a setter that would impact processing of the virtual keyboard and its interaction with the content. As is, it's really unclear what it's supposed to be doing. (e.g. I'm assuming you didn't mean for it to impact viewport changes that are unrelated to virtual keyboards)
• boundingRect should have a clear getter that defines what the returned values are.

/cc @gked

[snianu]

@yoavweiss Addressed all your comments here

[snianu]

This issue has been addressed in the linked PR.