https://bugs.webkit.org/show_bug.cgi?id=93052

Reviewed by Eric Carlson.

There was a regression in the Chromium port (http://crbug.com/139511) that revealed we didn't
have a layout test for suspend events. Upon further investigation it appeared we also had a bug
where we didn't fire the suspend event when a media engine reported they had completely loaded
the media.

Covered by new test http/tests/media/video-load-suspend.html.

Source/WebCore:

* html/HTMLMediaElement.cpp: Move firing of suspend event to changeNetworkStateFromLoadingToIdle
(WebCore::HTMLMediaElement::setNetworkState):
(WebCore::HTMLMediaElement::changeNetworkStateFromLoadingToIdle):

LayoutTests:

* http/tests/media/video-load-suspend-expected.txt: Added.
* http/tests/media/video-load-suspend.html: Added.
* media/event-attributes-expected.txt: Updated to include suspend event.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@125054 268f45cc-cd09-0410-ab3c-d52691b4dbfc