* fix depthwise conv gpu kernel bug, test=develop
* add more depthwise conv test, test=develop

* mem opt

* test=develop

* test=develop

* test=develop

* test=develop

* test=develop

* test=develop

* test=develop

* refine code  test=develop

* refine code  test=develop

* refine code  test=develop

* refine code  test=develop

* refine with cub test=develop

* fix mkldnn test && remove comments && test=develop

* polish code && test=develop

* add only_forward test && test=develop

* optimize depthwise conv by register memory
* test=develop

* refine reduce by cub
* optimize KernelDepthwiseConvFilterGrad
* optimize depthwise conv and reduce mean and reduce sum
* fix bug: dilation
* cuda arch and cuda 8 compatible

* "re-commit "

* "picked up"

* "fix ci"

* "fix pdb hang up issue in cuda 9"

* Fix more CPPLint issues

* Fix more CPPLint issues

* Fix more CPPLint issues

* Fix CPPLint issues in operators/math and operators/reader