Use std::string instead of WebString in WebRuntimeFeatures

Use std::string instead of WebString in WebRuntimeFeatures

The reason why I want to make this change comes from following facts:
- WebString's internal implementation is String, and String uses
  PartitionAlloc to acquire memory.
- PartitonAlloc must be initialized before any memory allocation.
  blink::initalize() does the job.
- On the other hand, runtime features needs to be enabled/disabled before
  blink::initalize() because Blink's initialization logic depends on the
  configuration. WebRuntimeFeature provides APIs to enable/disable
  features and it depends on WebString. Conflict happens.

By using std::string instead of WebString, we can avoid this problem
because std::string doesn't depend on PartitionAlloc.

We could take other options to fix this problem. For example, we can
separate initialization into 2 phases. I didn't take this approach
because it makes Blink API use be complicated.

BUG=501171
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=199972

[bashi, 2015-07-31 06:37:40]

Patchset #1 (id:1) has been deleted

[bashi, 2015-07-31 06:40:22]

bashi@chromium.org changed reviewers:
+ chrishtr@chromium.org, dglazkov@chromium.org, jochen@chromium.org,
pdr@chromium.org, philipj@opera.com, tkent@chromium.org

[bashi, 2015-07-31 06:40:22]

Hi API owners,

What do you think about this change?

I want to fix http://crbug.com/501171 and the root cause is that chrome side
calls/creates Blink's API/objects before blink::initialize(). Please refer the
description for details.

If this looks reasonable, I'll create 3-way CLs.

[jochen (gone - plz use gerrit), 2015-07-31 07:04:52]

looks reasonable

[tkent, 2015-07-31 08:20:44]

sounds reasonable to me too.

[haraken, 2015-08-04 00:15:54]

non-owner LGTM

[tkent, 2015-08-04 00:18:58]

Looks no one objects this.  lgtm.

[bashi, 2015-08-04 09:22:48]

On 2015/08/04 00:18:58, tkent wrote:
> Looks no one objects this.  lgtm.

Landed https://codereview.chromium.org/1269793007/ as the first series of
changes. I'm going to land this CL.

[bashi, 2015-08-04 10:18:51]

The CQ bit was checked by bashi@chromium.org

[commit-bot: I haz the power, 2015-08-04 10:18:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1259763006/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1259763006/20001

[commit-bot: I haz the power, 2015-08-04 11:37:28]

Committed patchset #1 (id:20001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=199972