remove (redundant) twopointradial shader

remove (redundant) twopointradial shader

1. Does anyone need the _BitmapType enum in SkShader anymore?
2. Why is there a compiler-assert in SkGpuDevice for the values?

client: use 2-point conic instead.

Abandoned : moved to https://codereview.chromium.org/1117423003/

BUG=skia:3800

[reed2, 2015-05-04 02:08:24]

The CQ bit was checked by reed@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-05-04 02:08:28]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1114243005/1

[reed2, 2015-05-04 02:15:54]

The CQ bit was checked by reed@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-05-04 02:16:03]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1114243005/20001

[reed2, 2015-05-04 02:16:33]

reed@chromium.org changed reviewers:
+ bsalomon@google.com, bungeman@google.com, djsollen@google.com,
halcanary@google.com, scroggo@google.com

[commit-bot: I haz the power, 2015-05-04 02:19:58]

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

[commit-bot: I haz the power, 2015-05-04 02:19:59]

Dry run: This issue passed the CQ dry run.

[reed2, 2015-05-04 11:09:49]

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

[bsalomon, 2015-05-04 14:09:14]

On 2015/05/04 11:09:49, reed2 wrote:

lgtm, the skgpudevice.cpp code is just legacy asserts that should be deleted. We
don't use that introspection any more.

[hal.canary, 2015-05-04 15:13:50]

The following PDFs render differently under pdfium:

*   blurrects.pdf
*   gradients.pdf
*   gradients_degenerate_2pt.pdf
*   gradients_local_perspective.pdf
*   gradients_no_texture.pdf

All but gradients_degenerate_2pt are okay minor differences. 
gradients_degenerate_2pt now renders correctly when it formerly failed.

PDF: LGTM

[djsollen, 2015-05-04 15:16:28]

kTwoPointRadial_BitmapType doesn't show up in android sources so...lgtm