Always use the native frame for popups and hosted apps on desktop linux

Always use the native frame for popups and hosted apps on desktop linux.

Previously, OpaqueBrowserFrameView was being used for popups and hosted apps. So popups and hosted apps would only use the native frame when the "Use system titlebars and borders" option was selected.

BUG=376137
TEST=Manual, see bug (Must use i3 window manager)

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=278392

[pkotwicz, 2014-06-11 03:08:50]

Elliot, can you please take a look?

This CL is an improvement over what's there but is not ideal.
We eventually want something similar to PopupNonClientFrameView.

The correct thing to do would be to either:
- Implement PopupNonClientFrameView::PaintToolbarBackground(), potentially
refactor PaintToolbarBackground() to get rid of duplication
- Change ToolbarView::PopupTopSpacing() such that popups with an OS frame behave
the same as the windows glass frame

I don't mind fixing the browser popup badness in my 20% time if you can talk to
people in MTV to come up with a spec of what popups on Linux should look like

[Elliot Glaysher, 2014-06-11 19:11:33]

I am actually fine with this. lgtm, but I'm ccing mgiuca, who might have some
opinion on this.

[pkotwicz, 2014-06-16 17:07:55]

mgiuca@ ping?

[Elliot Glaysher, 2014-06-16 22:53:42]

On 2014/06/16 17:07:55, pkotwicz wrote:
> mgiuca@ ping?

(If Matt is out, lgtm)

[Matt Giuca, 2014-06-17 02:57:34]

Hey, terribly sorry. I was out for 3 days last week and still catching up on
emails.

The change looks good (lgtm). I think V1 app windows should be considered
independently of the browser window setting.

Please fix the CL description, replacing "apps" with "V1 apps" everywhere. This
change doesn't affect V2 apps, which already always use the native frame style.
(Note that "V1 apps" refers to hosted apps, and the old deprecated packaged
apps, both of which use browser windows.)

[Matt Giuca, 2014-06-17 03:00:56]

Also, I'm a bit confused by the deletion of 3 files in this CL. They don't
appear to have been used prior to this CL (ie, you can delete them without
making the change to BrowserFrame and it still compiles and runs fine). If that
is indeed the case, it might be nicer to just make the change to BrowserFrame in
this CL, and make a separate CL deleting those files, called "Delete unused
files."

[pkotwicz, 2014-06-18 14:51:52]

I am leaving the deletion of the files in. The bug was partly because we thought
the files which this CL deletes were being used.

[pkotwicz, 2014-06-18 14:52:43]

The CQ bit was checked by pkotwicz@chromium.org

[commit-bot: I haz the power, 2014-06-18 14:53:03]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/pkotwicz@chromium.org/327083003/20001

[commit-bot: I haz the power, 2014-06-18 17:31:14]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-06-18 17:31:15]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/chromium_presubmit/bu...)

[pkotwicz, 2014-06-18 19:00:50]

James for OWNERS

[pkotwicz, 2014-06-18 19:02:30]

Scott for OWNERS

This is a non-CrOS change so I cannot use James.

[James Cook, 2014-06-18 20:32:23]

LGTM anyway :-)

[sky, 2014-06-19 15:05:05]

LGTM

[pkotwicz, 2014-06-19 15:44:16]

The CQ bit was checked by pkotwicz@chromium.org

[commit-bot: I haz the power, 2014-06-19 15:44:47]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/pkotwicz@chromium.org/327083003/20001

[commit-bot: I haz the power, 2014-06-19 16:16:35]

Change committed as 278392