* "unified operators"

* "add CUDNN register"

* "add use cudnn attribute"

* "add attribute"

* "test conv tranpose op"

* "remove duplicated attr"

* "fix op test"

* "add attribute to set cudnn"

* "add more log"

* "need layout op register support"

* "add more log"

* "change GetExpectedKernelType "

* "fix Get attr in conv_op"

* "fix CI"

* "fix tests"

* "removed kernel priority fallback"

* "fix CI"

* "fix stack pointer bug"

* "refine buggy interface"

* "add const cast to save life"

* "fix get_output_with_grad"

* "fix op test with dataformat"

* ""fix pooling

* "fix pooling test"

* "fix CI"

* "fix with_gpu error"

* "add transform needed functional check"

* "fix unpack list error"

* "comment out parallel.do temporary"

* "fix CI"

* "fix compile doc error"

* "make threshold larger"

* Add Seq2BatchFunctor, which will be used in WarpCTCOp.

* Implement WrapCTCFunctor and WrapCTCKernel.

* Add unittest of warpctc_op.

* Modify the check_output inferface in python unittest framework to allow check a subset of outputs.

* Use absolute offset lod in warpctc_op and related functors.

* Refine the comments of warpctc_op.

* The new python unittest supports checking a subset of the outputs, so revoke the previous change.

* Rename the transform from LoDTensor to Tensor with shape [max_sequence_length, num_sequences, sequence_width] to PaddingSequenceFunctor.

* Update to the newest codes.

* Rename the PaddingSequenceFunctor to PaddingLoDTensorFunctor and remove the computation of dimensions out of the functos.

* Rename Tensor::CopyFrom to Tensor::Copy

* Fix CI

* Fix compile

* "fix data transform"

* "data transformer"

* "add device pool"

* "add test"

* "fix ci"

* "fix datalayout implementation "

* "fix based on comment"

* implement selectedrows serialize and deserialize

* make serialize/deserialize as global function

* recover send_imp.cc

* delete unused brackets

* fix compile error

* serialize version in LodTensor and SelecetedRows

* fix ci

* fix ci

* remove unused place

* fix ci

* Remove unnecessary reshape in ColwiseSum

Speed up 12s -> 10s.

* Hand write ColwiseAdd in CPU

* add more place_test and rename Cudnn to CUDNN

* fix ci

* "derived cudnnDevice context"

* "leave remove cudnn handle from CUDADeviceContext"

* "fix math function error"

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`