* Update backward appending stragety to support double backward and fix some bug. (#18104)

* Update backward.py:
     - If there is no input grad var in all outputs of previous ops, do not append this op into graph.
     - Only apply this stragety when double backward.
* Update some double backward op.
* Update sum_op to judge whether a tensor is empty by numel or IsInitialized().

* add use_cuda to inplace pass,test=develop

* add test softmax_with_xe_inplace test,test=develop

if dim is -1, compile time check fails.

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

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

Fix code to support cpplint syntax check

test=develop

* fix GetTensorFromVar
test=release/1.1

* refine GetTensorFromVar
test=develop

* fix sum_op's GetExpectedKernelType
test=develop

* fix ci fail
test=develop

test=develop

This reverts commit 4d8e8ee2, reversing
changes made to d6a9f005.

* "add inplace attribute"

* "register inplace attribute"

* "change se-next model for memory-reuse"

* "fix typo"

* repick

* fix merge conflict

* "fix stupid error"

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.

* init

* add some check

* add dist transpile logic

* add insert op for block

* init change get_pserver_program

* optimize code

* fix a bug

* can run now

* start to do table split

* start to process table gradient

* complete pserver part

* can send_vars now

* revert cpplint

* fix a bug

* optimize code

* move dist test to models

* revert the interface of distribute_transpiler.transpile

* fix prefetch_block

* optimize trainspiler code

* add comment to sum_op

* add warning log

* fix comment

* fix test_send_recv

* fix test_send_recv

* fix train with no distributed table

* optimize GetDims

Fix sparse update memory error for distributed training

* Move Pod Types from DataType enum to Type enum

* Fixed data_type.h

* Fix type in TensorDesc

* Add comment to framework.proto

* Fixed type in data_type.h

* Updated format of type in data_type.h

* Fix var_desc.h

* Fix op_kernel_type.h

* Fixed data_type_transform_test.cc

* Fix operator.h

* Fixed data_type_transform.cc

* Fixed op_kernel_type_test.cc

* Fix operator.cc

* Fixed data_layout_transform_test.cc

* Fix var_desc.cc

* Fixed assign_value_op.cc

* Fixed assign_value_op.h

* fixed protobuf.cc

* Fix data_layout_transform_test.cc and op_kernel_type_test.cc

* Fixed rnn_memory_helper_op.cc

* Fix progrma_desc_test.cc

* Fixed fill_constant_batch_size_like_op.cc

* Fix operator_test.cc

* Fixed fill_constant_op.cc

* Fixed gaussian_random_op.cc

* Fixed uniform_random_op.cc

* Fixed edit_distance_op.cc

* Fixed fill_constant_batch_size_like_op.cc

* Fixed rnn_memory_helper_op.cc

* Fixed chunk_eval_op.cc

* Fixed assign_value_op.cc

* Fixed assign_value_op.h

* Fixed cast_op.h

* Fixed cast_op.h

* Fix fill constant op

* Fixed clang for assign_value_op.cc

* Fix one_hot_op.h

* Fix one_hot_op.cc

* Fix fill_op.cc

* Fixed sum_op.cc

* Fixed sum_op clang

* Fix uniform_random_op.cc

* Fix gaussian_random_op.cc

* Fix backward.cc

* Fix protobuf.cc

* Fixed prune_test.cc

* Fixed op_registry_test.cc

* Fix data_device_transform_test.cu

* Fix travis error

* Fixed one_hot_op.cu

* Fixed op_registry_test.cc

* Fixed nccl_op.cc

* Fixing python tests

* Revert "Fixing python tests"

This reverts commit fccaa4c5.

* Fixing Pybind to remove data type

* Fixing tensor.py

* Updated the new files:

* Resolve error in merge conflict of fill_constant_batch_size_like_op.cc

* Refine Type system

* Fixing type inference

* Fixed create_reader_op.cc

* Fix var_desc.h

* Fixed executor.cc

* Fix shape_inference.h

* Fixed create_reader_op.cc

* Fix tensor_util.h

* Fixed var_type_inference_test.cc

* Fix shape_inference.cc

* Fixed sum_op.c

* Fixed read_op.cc

* Fix var_type.h

* Fixed beam_search_decode_op.cc

* sendrecvop_utils.cc

* Fix operator.cc

* Fixed lookup_table_op.cc

* Fixed op_desc.cc

* Fixed get_places_op.cc

* Fixed lod_rank_table_op.cc

* Fixed beam_search_op.cc

* Fix var_desc.cc

* Fixed lod_tensor_to_array_op.cc

* Fixed while_op.cc

* Fix program_desc_test.cc

* tensor_array_read_write_op.cc

* Fix assign_op.cc

* Fix executor.cc

* Fix protobuf.cc

* Fix protobuf.cc

* add rename guard

* add device_data_transform

* add device_data_transform_test

* modify GetExpectedKernelType

* update operator.run

* support test test_label_semantic_roles

* optimize code

* optimize code

* rename GetActualKernelType to GetExpectedKernelType

* fix chunk_eval_op and device_data_transform_test

* add is_same_place to place

* optimize code, refine rename_guard

* refine rename guard, add GetKernelTypeForVar

* optimize code

* add some log

* rename guard

* use sub scope to create var

* fix compile

* add IsInitialized for Tensor

* add VarIsTensor

* fix op_registry_test

* test

* tmp disable priority

* restore switch_kernel.md

* code clean

* init kernel hint

* fix typo

* rm unused code

* add include in op_kernel.h

* restore op_kernel since it will be moved to op_kernel_type

* change force_cpu to use_cpu

* fix compilation

For input argument with a list of variables, drop_empty_grad is not allowed because it makes the correspondence bewteen a variable and its gradient ambiguous. Use REGISTER_OP_EX to register the op or call InputGrad(?,false) in GradOpDescMaker.

* Rename XXDescBind --> XXDesc

* Fix Compile

* 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`

* Add DataFeeder

A v2 API like data feeder for book demos.
We can feed data directly from reader.

* Fix CI

* Add an unittest for while/rnn op forward

* Add unittest for raw while op backward

* Fix CI

* Support int64 for sum op

* Refine code