WebInbandTextTrack should be deallocated by caller

WebInbandTextTrack should be deallocated by caller

An object that inherits from WebInbandTextTrack is created
on the Chromium side of the WebKit interface, and handed to
the InbandTextTrackPrivateImpl (via its constructor). The
InbandTextTrackPrivateImpl object currently assumes
ownership of the pointer, and so the
InbandTextTrackPrivateImpl object deallocates the
WebInbandTextTrack object (via its destructor).

However, this means that abstractions on the Chromium side
of the WebKit interface can potentially outlive the
abstractions on the WebKit side on which they depend.

To ameliorate such lifetime issues, the responsibility for
deallocating the WebInbandTextTrack object was moved to the
Chromium side, hence the InbandTextTrackPrivateImpl object
is no longer required to deallocate that object.

Note that the this change does not have a dedicated unit
test, because no code exists that actually calls this code.
These changes to WebKit are a prerequisite for the change
that adds support for inband text track rendering to Chrome,
as enunciated here:

https://chromiumcodereview.appspot.com/13419002/

The requisite infrastructure for testing all of the changes
in toto will be included when that cited change is
committed.

BUG=230708
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=150686

[Matthew Heaney (Chromium), 2013-05-16 22:03:41]

This is a small change to the WebKit side of things.  The lifetime of the
WebInbandTextTrack object is now controlled on the Chromium side.

[acolwell GONE FROM CHROMIUM, 2013-05-17 16:27:37]

lgtm

[Matthew Heaney (Chromium), 2013-05-17 22:26:05]

Tentatively the final change to WebKit before landing the CL for inband text
tracks.

https://codereview.chromium.org/13419002/

[abarth-chromium, 2013-05-20 17:44:35]

Ok...  I don't feel like I have enough context to verify that this CL is
correct, but I believe you.  Would you be willing to write a comment in one of
the API headers about the ownership relations among these various objects? 
(Feel free to do that in a followup CL.)  LGTM

[commit-bot: I haz the power, 2013-05-20 17:48:27]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/matthewjheaney@chromium.org/15236002/1

[commit-bot: I haz the power, 2013-05-20 18:46:15]

Change committed as 150686