NotificationObjectProxy should own the DesktopNotificationDelegateImpl

NotificationObjectProxy should own the DesktopNotificationDelegateImpl

This makes the memory model more robust for tests (which may not interact
with a notification), and means that the NotificationObjectProxy won't be
stuck with a bad pointer after the delegate deletes itself.

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=287518

[Peter Beverloo, 2014-08-02 16:51:06]

+Justin for notifications
+Jochen for render_frame_host_impl.cc OWNERS

I've enabled the TestNotificationReplacement test so that the LSan and ASan bots
will verify its correctness. Locally [AL]San builds are happy, and manually
counting live objects in the constructor and destructor and then interacting
with the notifications works well too. The Proxy itself is stored as a scoped
refptr.

[jochen (gone - plz use gerrit), 2014-08-04 09:11:24]

https://codereview.chromium.org/433273002/diff/1/chrome/browser/notifications...
File chrome/browser/notifications/notification_object_proxy.h (right):

https://codereview.chromium.org/433273002/diff/1/chrome/browser/notifications...
chrome/browser/notifications/notification_object_proxy.h:28:
content::DesktopNotificationDelegate* delegate);
you could make the parameter a scoped_ptr<> to make it clear that it's
transferring ownership

[Peter Beverloo, 2014-08-04 11:33:34]

+benm for android_webview/

https://codereview.chromium.org/433273002/diff/1/chrome/browser/notifications...
File chrome/browser/notifications/notification_object_proxy.h (right):

https://codereview.chromium.org/433273002/diff/1/chrome/browser/notifications...
chrome/browser/notifications/notification_object_proxy.h:28:
content::DesktopNotificationDelegate* delegate);
On 2014/08/04 09:11:24, jochen wrote:
> you could make the parameter a scoped_ptr<> to make it clear that it's
> transferring ownership

Done. Plumbed through the ContentBrowserClient as well so that it's clear across
the layers.

[jochen (gone - plz use gerrit), 2014-08-04 12:04:47]

lgtm

[benm (inactive), 2014-08-04 13:04:14]

aw lgtm

[dewittj, 2014-08-04 20:51:12]

lgtm, thanks!

[Peter Beverloo, 2014-08-05 10:14:05]

The CQ bit was checked by peter@chromium.org

[commit-bot: I haz the power, 2014-08-05 10:15:11]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/peter@chromium.org/433273002/20001

[commit-bot: I haz the power, 2014-08-05 11:19:41]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  chromium_presubmit on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)
  linux_chromium_chromeos_rel_swarming on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[commit-bot: I haz the power, 2014-08-05 12:04:06]

Change committed as 287518