Refine the GlobalEventHandlers test

Refine the GlobalEventHandlers test

http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2013-October/thread.html#40980

The conclusion of the WHATWG "Should onfoo event handler properties be
on Element or HTMLElement?" thread was that not letting Element have
these properties is more sane, because reflecting the corresponding
content attributes for Element would be odd.

Update the test to assert that Element does not have the properties,
but that HTMLElement and SVGElement do.

Also split the enumerate tests so that individual results can be seen.

BUG=305112
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=159478

[philipj_slow, 2013-10-11 10:36:24]

For some context, here's my current plan for GlobalEventHandlers:

1. Update the test (this review)
2. Remove PerWorldBindings
3. Remove NotEnumerable
4. Add the missing attributes

At this point the lists of attributes in Element, Document and Window should be
in sync. All tests should be passing except the "No Element.onfoo" and those for
events not implemented yet.

5. Move to a shared GlobalEventHandlers IDL for HTMLElement, SVGElement,
Document and Window

Now the "No Element.onfoo" tests should pass.

6. Refactor the C++ bits to share as much code as possible.

This should make the change easy to bisect, when some breakage from real-world
content (probably) happens.

[abarth-chromium, 2013-10-11 18:22:21]

lgtm

[commit-bot: I haz the power, 2013-10-11 18:22:34]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/philipj@opera.com/26478004/1

[commit-bot: I haz the power, 2013-10-11 22:14:37]

Change committed as 159478