Also add `elemwise_sub/mul/abs/clip` layers

Fix #6531

typo WITH_TEST to WITH_TESTING

* "add registry interface"

* "move function to registry"

* "rename with meaningful name"

* "add exposed layers"

* "fixed based on comments"

* "remove unsed comments"

* first commit

* Update read_source.md

* fix the ending symbol

* fix invalid content link

add prelu neon impl

fix warning in row_conv_op.cu

* Add fill_op

* Fix bug

merge Release/0.11.0 to develop

* update docs about how to add a new operator

There are mainly following fixes:

- take `DeviceContext` as the template parameter of math functors and OpKernel instead of `Place`
- remove `eigen_device` interface in base class  `DeviceContext`
- remove `GetEigenDevice` interface in `ExecutionContext` and base class `DeviceContext`
- remove unused `platform::EigenDeviceConverter`
- rename `REGISTER_OP_GPU_KERNEL` to `REGISTER_OP_CUDA_KERNEL`
- rename `USE_GPU_ONLY_OP` to `USE_CUDA_ONLY_OP`

fix img_pool maxout doc

Fix wrong index in dataset downloading exception

skip mkl setting in v1 with Mac

Use PADDLE_WITH_NATIVE_FP16 for float16_t.

* Updating the Latex equation for Adagrad

* Fixing Latex euqations for adadelta, adam and adamax

* Fix documentation

* Address review comments

* "add global regularization"

* Polish `append_regularization_ops`

Update annotations of layers.py

recv_op use serialized program

unify MKL macro definition

* Fix compiling error of gcc4.9.

* Refine the check of cxx compiler flags in api/CMakeLists.txt.

add inference benchmark data