* add split_ids_op

* add TestSplitIdsOp

* fix comment

* add test for empty tensor

* clean code

* rm unused code

Add pinned memory

 add MKL for fluid static and shared library

Fix test_recv_op

* add LRN layer for fluid

* add LRN layer for fluid

* add documentation for LRN layer

* add paper reference for LRN layer

* add seperate documentation for LRN layer

* rm lrn.py in doc/fluid/dev/src

* change code style in lrn

* fix style of comments in lrn

Improve layer_norm speed

Split send op to send_vars and send_barrier

MKLDNN Relu Tanh Sqrt Abs activations added

* support empty tensor

Fix links in the English Documents

Fix concat_op

"fast hack"

    transfomer on a single device step time
    reduces from 0.157 to 0.125

fix dist compile