Rewrite apps::SavedDevicesService as extensions::DevicePermissionsManager.

Rewrite apps::SavedDevicesService as extensions::DevicePermissionsManager.

This service doesn't need any Chrome dependencies after all and so it
can move into //extensions. This will make it possible for the USB
extensions API code to directly depend on it. In the process I have
rewritten the service with a couple improvements:

* C++11 for-each loops are used instead of explicit iterators.
* The DevicePermissions object is no longer an inner class (so it can
  be forward declared) and it is explicitly copied when requested so
  that the ownership of data on the FILE and UI threads is clear.

BUG=

Committed: https://crrev.com/bc82d3c46cedec68b175c9572a7de412eb0e6427
Cr-Commit-Position: refs/heads/master@{#297230}

[Reilly Grant (use Gerrit), 2014-09-25 04:59:00]

Patchset #1 (id:1) has been deleted

[Reilly Grant (use Gerrit), 2014-09-25 05:32:01]

Patchset #1 (id:20001) has been deleted

[Reilly Grant (use Gerrit), 2014-09-25 05:35:40]

reillyg@chromium.org changed reviewers:
+ scheib@chromium.org

[Reilly Grant (use Gerrit), 2014-09-25 06:05:30]

Patchset #1 (id:40001) has been deleted

[Reilly Grant (use Gerrit), 2014-09-25 06:06:10]

This patch depends on https://codereview.chromium.org/601073002/.

[Reilly Grant (use Gerrit), 2014-09-25 23:13:35]

Patchset #1 (id:60001) has been deleted

[Reilly Grant (use Gerrit), 2014-09-26 20:54:06]

reillyg@chromium.org changed reviewers:
+ yoz@chromium.org

[Reilly Grant (use Gerrit), 2014-09-26 20:54:48]

Updated this patch with the latest C++11 goodness and added yoz@ to review the
extensions/browser/DEPS change.

[Yoyo Zhou, 2014-09-26 21:04:44]

On 2014/09/26 20:54:48, reillyg wrote:
> Updated this patch with the latest C++11 goodness and added yoz@ to review the
> extensions/browser/DEPS change.

I think you actually need a device/usb OWNER.

[Reilly Grant (use Gerrit), 2014-09-26 21:05:26]

On 2014/09/26 21:04:44, Yoyo Zhou wrote:
> On 2014/09/26 20:54:48, reillyg wrote:
> > Updated this patch with the latest C++11 goodness and added yoz@ to review
the
> > extensions/browser/DEPS change.
> 
> I think you actually need a device/usb OWNER.

I'm a device/usb OWNER.

[Yoyo Zhou, 2014-09-26 21:21:58]

On 2014/09/26 21:05:26, reillyg wrote:
> On 2014/09/26 21:04:44, Yoyo Zhou wrote:
> > On 2014/09/26 20:54:48, reillyg wrote:
> > > Updated this patch with the latest C++11 goodness and added yoz@ to review
> the
> > > extensions/browser/DEPS change.
> > 
> > I think you actually need a device/usb OWNER.
> 
> I'm a device/usb OWNER.

Ah, ok. Well, you don't need my LGTM, but that deps change LGTM.

[Reilly Grant (use Gerrit), 2014-09-26 21:22:36]

On 2014/09/26 21:21:58, Yoyo Zhou wrote:
> On 2014/09/26 21:05:26, reillyg wrote:
> > On 2014/09/26 21:04:44, Yoyo Zhou wrote:
> > > On 2014/09/26 20:54:48, reillyg wrote:
> > > > Updated this patch with the latest C++11 goodness and added yoz@ to
review
> > the
> > > > extensions/browser/DEPS change.
> > > 
> > > I think you actually need a device/usb OWNER.
> > 
> > I'm a device/usb OWNER.
> 
> Ah, ok. Well, you don't need my LGTM, but that deps change LGTM.

Presubmit thinks that I do. :/

[scheib, 2014-09-29 15:30:53]

lgtm with a few followups:
- Patch comment should describe the changes beyond the move.
GetPermissionMessageStrings, C++11 for each loops.
- Class comments

Code Review & later archeology is easier/faster when patches are doing fewer
things at once. Especially with moves of files it is much easier if you keep
other changes that will make the diff harder to read to another patch. This
patch is sort of borderline, around 400 lines of code changed enough that you
can't see what is happening in the diff and needing full read-through to
understand.

https://codereview.chromium.org/606503002/diff/100001/extensions/browser/api/...
File extensions/browser/api/device_permissions_manager.h (right):

https://codereview.chromium.org/606503002/diff/100001/extensions/browser/api/...
extensions/browser/api/device_permissions_manager.h:57: class DevicePermissions
{
http://google-styleguide.googlecode.com/svn/trunk/cppguide.html#Class_Comments
and classes below as well.

[Reilly Grant (use Gerrit), 2014-09-29 19:21:15]

The CQ bit was checked by reillyg@chromium.org

[commit-bot: I haz the power, 2014-09-29 19:23:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/606503002/120001

[commit-bot: I haz the power, 2014-09-29 19:28:02]

Committed patchset #3 (id:120001) as 5b5facd07b5b223f57d38ef7197a9b8e02fa1181

[commit-bot: I haz the power, 2014-09-29 19:28:45]

Patchset 3 (id:??) landed as
https://crrev.com/bc82d3c46cedec68b175c9572a7de412eb0e6427
Cr-Commit-Position: refs/heads/master@{#297230}