Revert of Added <link rel=preconnect> crossorigin attribute

Revert of Added <link rel=preconnect> crossorigin attribute (patchset #2 id:20001 of https://codereview.chromium.org/1138743002/)

Reason for revert:
The patch seems to be causing preconnect related crashes: https://build.chromium.org/p/chromium.webkit/builders/Linux%20ChromiumOS%20Tests%20%281%29/builds/13855

Reverting until I get a clearer picture as to why that is the case.

Original issue's description:
> Added <link rel=preconnect> crossorigin attribute
> 
> In order to support preconnect for both anonymous and non-anonymous connection pools we need to add support for the crossorigin attribute in <link rel=preconnect>.
> This CL implements the core functionality of this change on the Blink side.
> 
> BUG=468005
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=195184

TBR=mkwst@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=468005
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=195200

[Yoav Weiss, 2015-05-11 20:56:42]

Created Revert of Added <link rel=preconnect> crossorigin attribute

[commit-bot: I haz the power, 2015-05-11 20:57:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1131513004/1

[Dirk Pranke, 2015-05-11 20:57:35]

dpranke@chromium.org changed reviewers:
+ dpranke@chromium.org

[Dirk Pranke, 2015-05-11 20:57:36]

lgtm

[commit-bot: I haz the power, 2015-05-11 20:57:46]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=195200

[Mike West, 2015-05-12 03:10:59]

LGTM. Thanks for reverting. Happy to look at a second attempt when you have a
fix.

[Yoav Weiss, 2015-05-12 07:05:03]

On 2015/05/12 03:10:59, Mike West wrote:
> LGTM. Thanks for reverting. Happy to look at a second attempt when you have a
> fix.

I see the crash locally, and see what change caused it. It's the fact that we
temporarily broke the preconnect functionality that this test is testing (which
causes it to time out). I'll resubmit without calling the new interface for
preconnect (so without breaking the preconnect functionality)