remove (unused by chrome) SkDeferredCanvas(device) factory signature

remove (unused by chrome) SkDeferredCanvas(device) factory signature

BUG=skia:
R=junov@chromium.org

Committed: https://code.google.com/p/skia/source/detail?r=13302

[reed1, 2014-02-03 20:37:47]

still looking at benchmark.cpp, which is a little too SkDevice-crazy

[Justin Novosad, 2014-02-03 21:33:47]

https://codereview.chromium.org/136753016/diff/1/tests/DeferredCanvasTest.cpp
File tests/DeferredCanvasTest.cpp (right):

https://codereview.chromium.org/136753016/diff/1/tests/DeferredCanvasTest.cpp...
tests/DeferredCanvasTest.cpp:471: #if 0   // TODO -- need a surface-oriented way
to exercise this.
You could register an SkDeferredCanvas::NotificationClient and verify that
flushedDrawCommands() get called on the notification client after the 4th
drawBitmap.

[reed1, 2014-02-03 21:52:39]

ptal

https://codereview.chromium.org/136753016/diff/1/tests/DeferredCanvasTest.cpp
File tests/DeferredCanvasTest.cpp (right):

https://codereview.chromium.org/136753016/diff/1/tests/DeferredCanvasTest.cpp...
tests/DeferredCanvasTest.cpp:471: #if 0   // TODO -- need a surface-oriented way
to exercise this.
On 2014/02/03 21:33:47, junov wrote:
> You could register an SkDeferredCanvas::NotificationClient and verify that
> flushedDrawCommands() get called on the notification client after the 4th
> drawBitmap.

Done.

[Justin Novosad, 2014-02-03 22:28:09]

lgtm

[reed1, 2014-02-04 15:34:16]

Committed patchset #2 manually as r13302 (presubmit successful).