* add batch_norm_layer

* add img_conv_group layer and test

* add check to Tensor.type()

* forward can run

* with backward

* change label data time from int32 to int64

* refine code

* follow comment

* 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

* Remove template parameter for Tensor methods

* Also check the type is correct when data()
* Simplize holder_

* Fix accuracy_op

* Register Code

* Feed/Fetch op just plain operator, not a OpWithKernel
* Do not register OpInfoMaker since Feed/Fetch will never be
  configured by users
* Feed/Fetch op has empty gradient
* Feed/Fetch op do not hard code `feed_variable`, `fetch_variable` as
  its input and output, make it as a plain Operator input/output

* init

* unify CopyFrom interface

* fix gpu build error

* fix bug in tensor_py.h

* refine code comments and add TODO list

* fix conflicts in FeedOp and FetchOp

* Adding implementation for copying a vector to tensor
* Changing Tensor test to access gpu memory indirectly

* It is duplicated with `dim_`. We can use `dim_` to calculate `numel`
  everytime. It does not cost too much.
* `numel` is not initialized by constructor. Also, `numel` is hard to
  synchronize with `dim_`.

So just remove it.

* add lodtensor

* add reshape of lod

* add details

* rename Elements/Levels

* size_t and vector reserve

* add details

* add const& std::shared_ptr

* add lod_tensor_impl.h

* remove a shared_ptr

* fix some enforce

* remove compatible_type to avoid compile error

* remove shared_ptr

* fix tensor error msg

Fix #3171

1. Change PODDeleter's template parameter 'PlaceType' to 'Place'.

2. Limit PODDeleter and Tensor::mutable_data()'s `T` to POD type.

ATTENTION: some interfaces changed:
1. void Tensor::set_dims(const DDim& dims) ==> void Tensor::Resize(const DDim& dims).
2. void Tensor::ShareDataFrom(const Tensor& src)  ==> void Tensor::ShareDataWith(const Tensor& src)
3. DDim Tensor::dims() const ==> const DDim& Tensor::dims() const