* add elementwise div

* move mul and div grad functor

* Combine multiple CUDA kernels

* Update the reduce interface call

* add multi-output

* add multi-output div

* add branch judge

* Package branch

* Combine the x and y functions into one

* change 'math' to 'math_kernel'

* fix compile bugs

* merge develop

* fix compile bugs

* Reduce reshape kernel functions in pten

* delete notes

* fix bugs when compile

* modify register name

* fix compile bugs

* elementwise_div refactor

* fix compile bugs in windows ci

* modify complex template for elementwise ops

* modify mul, div grad struct

* add complex template for CudaShuffleDownSync CudaShuffleXorSync funcs and fix the bug when delete cuda<9000

* fix shuffle func args bug

* fix shuffle func args bug

* fix shuffle func args bug

* [ROCM] update fluid elementwise op for rocm (part10), test=develop

* update, test=develop

* address review comments, test=develop

* More precise mkldnn kernel choice in GetExpectedKernelType

* Fixes after review

* Refresh develop for CI

* CI experiment

* get back from CI exper

* type promotion for grad

* add type promotion for div op

* change the grad of div when complex types

* fix the grads of inputs args order not match bug

* replace part of the old implementation, test=develop

* restore concat op, test=develop

* update all ops implemention & delete GetDataTypeOfVar func, test=develop

Improve elementwise operators performance in same dimensions

* update elementwise double grad to save gpu memory, test=develop

* update elementwise_mul/div_grad_grad to save memory, test=develop

* remove eval function in eigen statement to save memory, test=develop

* add unittest for elementwise_div_grad_grad without dout, test=develop

* add unittest for elementwise_add_grad_grad without ddx, test=develop

* add float16 cuda kernel for elementwise double grad op, test=develop

* double backward, elementwise_div

* fix dx empty. test=develop

* bug fix (#17392)

fix secure bug

* Eanble stack operator for a Ngraph, test=develop (#17406)

* fix sqrt_grad_grad unittest. test=develop (#17410)

* fix sqrt_grad_grad unittest. test=develop

* disable sqrt_grad_grad unittest. test=develop

* test=develop, fix unittest

* test=develop, fix unittest

* test=develop, fix unittest

* test=develop, fix bug

* fix unittest. test=develop

* fix unittest dx. test=develop

* tmp fix! for test... test=develop

* reduce tmp, test=develop

* test=develop, reduce tmp

* fix broadcast unittest. test=develop

* fix format. test=develop

* refine code. test=develop

* refine code. test=develop

* refine GetDoubleGradSafeTensor. test=develop

* fix format. test=develop

* fix some grad op desc maker
test=develop

* fix grad op desc makers
test=develop

* wip simplify operator framework

* wip

* wip

* done test=develop

* clean test=develop

* fix test=develop

* fix deps test=develop

* fix cpu build test=develop

* fix tensorrt build test=develop

* fix tests test=develop

* fix test=develop

* fix cpu build test=develop

* fix input<tensor>
test=develop

* fix split_ids
test=develop

* ElementwiseMul should not support SelectedRows

* fix scale op
test=develop

* change GetTensorFromVar() method to GetTensorOrSelectedRowsFromVar()

* fix operator

* refine MultiOutput

* fix MultiOutput
test=develop

* disable test_dist_save_load
test=develop

* fix elementwise_op
test=develop

* add get_sparse_as_op
test=develop

* add info for check
test=develop

* rename get_sparse_as_op with extract_rows_as_op.
test=develop

* elementwise doesn't support selected_rows

* fix regularizer

* remove extract_rows_as
test=develop

* fix ci
test=develop

* add test for sum_op

* fix regularizer
test=develop

*  test=develop

* fix pserver weight decay multi inputs test=develop

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`

Elementwise operator add/sub/mul/div