Also add `elemwise_sub/mul/abs/clip` layers

* Add fill_op

* Fix bug

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`

* Updating the Latex equation for Adagrad

* Fixing Latex euqations for adadelta, adam and adamax

* Fix documentation

* Address review comments

* Fix compiling error of gcc4.9.

* Refine the check of cxx compiler flags in api/CMakeLists.txt.

* Fix bug in MergeModel.cpp.

* Add a c-api inferface to initilize the thread environment of Paddle and add a GPU example.

* Add some note for paddle_init_thread and move the inplementation of paddle_error_string into a .cpp file.

* Add some comments.

* init decoder_trainer

* can run

* fix lod

* add sharelod to cross_entropy_grad_op

* add avg_cost to fetch list

* modify learning rate

* can run

* optimie code

* add early exit

* fix print

* revert test_understand_sentiment_conv.py

* add act to fc

* fix gpu memory allocation policy

* refine codes

* fix code style

* follow comments

* Add HasCUDNN to detect if CUDNN is installed or not

* Fix CI

Add LRN efficient GPU implement

* 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

* Complete Dynamic RNN