Second attempt to land the integral image scaling change.

Second attempt to land the integral image scaling change.

Scale all images to the nearest rounded integer, and if there's still
any scaling factor left over, pass it on to the subsequent bilerp code.
Should avoid artifacts when tiling scaled images.

Original CL received an LGTM from reed; new version disabled tiling
in the downsamplebitmap GM; I verified that this fixes the issue
we were seeing there on non-neon androids.

BUG=skia:2888
TBR=reed

Committed: https://skia.googlesource.com/skia/+/d73c169637f1ab0b96bae43269da7216bb93a7e1

[humper, 2014-08-28 21:08:50]

The CQ bit was checked by humper@google.com

[commit-bot: I haz the power, 2014-08-28 21:10:00]

CQ is trying da patch. Follow status at
 https://skia-tree-status.appspot.com/cq/humper@google.com/514383003/1

[commit-bot: I haz the power, 2014-08-28 21:27:45]

Committed patchset #1 (id:1) as d73c169637f1ab0b96bae43269da7216bb93a7e1

[reed1, 2014-08-28 21:36:09]

reed@google.com changed reviewers:
+ reed@google.com

[reed1, 2014-08-28 21:36:09]

I presume this is a sort of diagnostic land, to validate that certain failures
will now be fixed, but do we understand the perf regression from the original
land?