Merge branch 'dist_train_benchmark_vgg16' of https://github.com/typhoonzero/Paddle into dist_train_benchmark_vgg16

Merge branch 'dist_train_benchmark_vgg16' of https://github.com/typhoonzero/Paddle into dist_train_benchmark_vgg16

* Updating the cluster trainign doc

* Fixed few comments of transpiler

* Adding few explanations

* Updating the cluster trainign doc

* Fixed comments

* Updating few comments in recv_op

update gradient clip api

compile and install the shared library of fluid inference

remove unused v1 api doc

Add python wrapper for  CTC greedy decoder and edit distance evaluator

Fix CI

Add v2 get layer out in FAQ

2. Fix evaluator using 'reduce_sum' op instead of 'mean' op

fix copyright

Move registery.py to layers module

Make compare_op reuse elemwise_op_funcs