* Add MKLDNN layout support in Paddle

Add MKLDNN layout in Paddle so that MKLDNN friendly memory layout
can be used in MKLDNN enabled OP kernel. Before this commit, NCHW
is hardcode to be used in all MKLDNN op kernels. As a result,
non-optimized execution path is selected in MKLDNN primitive which
bring worse performance.
Besides framework change, three MKLDNN OP kernels were updated
for using new MKLDNN layout. They are conv/pool2d/batch_norm.
Other MKLDNN OP kernels need be also updated in similar way to
achieve best performance.

* Add MKLDNN layout support in activation OP

* Don't populate layout from input to output when kMKLDNN in

* Refine pool mkldnn op kernel

* MKLDNN layout

* Remove the inferitance from tensor file

* MKLDNN layout: refactoring

* Remove additional #define to register new operator

* Prepare mkldnn tests to work with layout

Do not use ctor

* Reduce line of codes.
* We can use virtual function for Maker now.
* The implementation does not care what maker holds, it is easier to
refactor later.

* script to add semicolon

* fix typo

* Move framework.proto to proto namespace

* Fix compile

* Fix compile

* Fix Compile

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`

* Using latex fraction syntax in sigmoid and logsigmoid op

* Fixing the rendering of the latex equations in adamax operator

* "add floor, ceil, round op"

* "reuse zero gradient"

* "fix divide zero"

* "fix numpy floor error"

* Support many data types of several operators

* SeqConv only support float/double

* Revert adagrad

* fix m_ops

* fix activation op

* Simplize Gradient Check

* Stash

* Extract apply_backward_pass to backward.py

Rename apply_backward_pass to append_backward_ops

* Use graph API to check gradient

* Fix ci

* Fix CI

* Fix backward for double precision

* Stash

* Fix CI

* Fix ci

* Ignore GRU test

* Ignore xe op

* Fix CI

* Fix softmax with xe gradient

The correct equation should be IG = OG * (d_softmax_with_xe())

* Fix typo

* Fix merge error

* Disable LRN

They are public now

* Adding Hard Sigmoid Activation

* Adding a comment for slope to be only positive

* Fixing grammatical mistake in comment

* Adding thresholded_relu op
* Adding test for thresholded relu op

* implementing softplus

* small fix

* small fix

* small fix

* small fix

* Implemented the hardShrink activation

* Fixing the unit test

* Add numerically-stable logsigmoid activation

* Add softshrink operator

* Adjust relative tolerance for grad-check

* Address review comments