1. Add template T which indicates data type to `CopyFrom()`, `Slice()`
and `ShareData()` functions. This makes `CopyData()` code much clearer.

2. Add `set_dim()`.

3. `product(DDim)` transforms `DDim` to `vector<int>` first and then calculate
its product. That might be quite slow. For `product(dims_)` is frequently
used in Tensor, we add a mumber variable `numel_` as a cache of the
product result.
TODO: refactor `product()` to make it more efficient.

4. Unable Tensor::operator=

5. Remove the limit of POD type, because `float16` and `int8` are not POD type.

Add a sample op, `add_op`

* Convert `op` --> `operators`
* Remove AddType in OpProtoMaker, because type is part of registry.
* Rename CPU_OR_GPU --> DEVICE_TYPE in registry macro.

Default scope function

* Refine register methods, make Op can get rid of whole-archieve
* `USE_OP` before a op is used.
* Add unittest for add_op.

fix bug in dynload

Compile for armv8.

"fix init error"

go_binary: remove hardcoded library link path, add pserver client test

* Server recovery from checkpoint

Add go testing into cmake

Add unit test for OpKernel

Add GetEigenDevice To DeviceContext.

FIX: add -lrt for link

1. Add `Tensor::CopyFrom`. Current version can only support CPU memory
copy. The support of GPU will be provided later by `paddle::memory`.
The current implementation of `Tensor::CopyFrom` is a little inefficient:
Every time `CopyFrom` is called, tensor will re-allocate its memory. However, if
we try to check and reuse `placeholder_`, we have to provide a template
parameter for `CopyFrom` to indicate the data type. It seems strange for
a simple copy function.

2. Add `Tensor::mutable_data(Place place)`, which directly use member
variable `dims_` as its dim parameter. This interface is required by
`Op::InferShape`.

fix pybind compile question