Node should be created by Graph::CreateXXX so that
    they are managed by graph.

* remove template operator

* "fix compile"

* "fix ci"

* "fix ci"

* fix dist train reduce mode

* fix previous fix

* refine dist train RPC deps

* clean up

* clean up

* fix ut

* remove input for fetch_barrier

* follow comments

* dist transpiler add control dependency var between send and recv

* fix async deps

* follow comments and refine

* fix deps connect for rpc ops

Reduce one level of inheritence.

* fix the order of sum

* add doc

* check whether need to copy

* follow comments

Add a few more RecordEvent.
Cleanup

fix fluid_benchmark.py

add doc