neural machine translation model support beam search with while op

* Fix ParallelDo not support empty input gradient

* Polish ParallelDo and fix several bugs

* Fix CI

* Fix CI

* add GetLoD for debug

* add LoDToString

* optimize if

* typo

* add lod_tensor to operator's dependency

feature/parallel_gpu

* Rename Tensor::CopyFrom to Tensor::Copy

* Fix CI

* Fix compile

* Remove unused LoDTensor methods

* Update

* send_recv variable

* delete unused logs

* fix ci failed

* update

* resize tensor before tensor copy

* add selectedrows unit test

* check rows

* 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

* Move framework.proto to proto namespace

* Fix compile

* Fix compile

* Fix Compile

* WIP send recv op

* WIP send recv

* put grpc impl in details

* put grpc impl in details

* update wip

* update proto

* update proto

* update proto

* clean cmake

* wip on op implementations

* wip on op implementations

* compile ok adding ut

* wip unitest

* add extern cares for linking

* wip add ut

* working version send recv

* revert optimizer.py

* update test cmake

* add libtool to dockerfile

* update cmake dependency

* update cmake depends

* update cmake grpc depends

* fix cmake dependency

* fix compile error

* fix compile

* follow comments

* update

* update copyfrom

* Add LoDRankTable

LoD Rank Table stores the `level` of `lod` which is ordered by sequence
length in descending order. It is useful when implement dynamic RNN and
is shared by dynamic RNN memory, dynamic RNN slice input and dynamic
RNN slice output operators.

* Add skeleton for array_to_lod_tensor and lod_tensor_to_array

* Add VarType::LoDTensorArray

* Add PyBind of LoDTensorArray

* Add InferVarType

* Add first unittest

* Add ut

* Add unittest

* Add unittest

* Add unittests

* update

* init

* add infershape for lod_tensor_to_array_op

* compelete array_to_lod_tensor_op

* copy data

* clean code

* clean code

* Fix unittest data

* fix bugs

* fix compile error

* Refine TensorToArrayOp

* refactor array_to_lod_tensor

* Unittest

* fix bugs

* Fix unittest

* Fix unittest

* debug

* Debug

* Fix unittest

* clean code

* refactor

* use ostream

* update test

* fix gpu build error

* make gpu test pass

* Add GetFineGrainedLoDLength and AppendLoD

* Follow comments and fix bugs

* fix a compile error

* fix a compile bug

* Init

* Stash

* Polish SaveLoadOp

* Fix CI

* Polish code

* Save GPU Tensor

* Stash

* Fix CI

* "add model format design doc"

* "add restore function"

* "add parse protobuf"

* "move necessary information to saver.proto"

* "format code"

* "add gpu option"

* "add lod info"

* "add saveop python test wrapper"

* "checkpoint reuse save operator"

* "rewrite model format design doc"

* "async support needed"

* "fix run once"

* "fix doc based on comments"

* "refine based on comments"

* "fix based comments"

* "remove persistable flag from framework.proto"

* "add IndicateDataType to restore op"

* "add save test"

* "modify save restore code"

* "modified the restore logic"

* rm checkpoint_op.cc

* rm test_checkpoint_op.py

* "get inputs outputs name from execution context"

* Saving each variable to a independent file

* Fix bugs

* Rewrite save_restore_op_test with new Python framework

* Move `SaveOp` and `RestoreOp` from OpWithKernel to OpBase

* Refine unit test of SaveOp and RestoreOp

* fix compile errorwq

1. Add more comments and exmples
2. Rename repeat_lod to expand_lod
3. Remove unused head file

* some enforce change

* add LoDTensor::NumElements(id,id)

* update

* init

* finish

* refacter interfaces

* remove SliceCopied

* remove SliceCopied

* rename SliceShared to SliceLevels, SliceInLevel

* merge lod_tensor/details

* remove lod_start_pos_'s shared_ptr

* make lod-tensor a special tensor

* add clone to lod_tensor

* add lod tensor impl

* add lodtensor clone test

* init lod

* add equal

* merge LOD and its methods

* recover tensor and variable

* change thrust to host_vector