* dist transpiler add control dependency var between send and recv

* fix async deps

* follow comments and refine

* fix deps connect for rpc ops

* Add Python Callstacks when Op::Run error

* Skip op with sub-block

* refactor: refine callstack info's format

* Reshape only support matrix

* Polish Python code

* Fix UT

* Fix Py3

* Disable in_place in batch_norm API.

* Fuse Convolution and Eltwise Add into Conv+Bias

* Reduce bias branching at conv_mkldnn_op

* Add MKLDNN build checks for Conv Bias

* Conv-bias: check if bias input exist befor assignment

* Conv-bias: Remove Bias dim check from infershape

It was causing conv3d test to crash upon\ncalling HasInput(Bias)

* add async dist tests

* update delta

* fix transformer test

* refine rmsprop transpile

* update

* fix dist seresnet

* Fix scatter_op python API and remove inconsistency between implementation and doc

* API spec change

* Change as review comment

Port dist_transpiler new added codes
Port ut for clone desc

* Fix bug in conditional_block_op.
* Fix bug and add comments.
* Rename arguments.

Add dependency to send recv

Optimize selected rows for dist lookup table with rwlock 

* remove patchelf in windowls

* "follow comment"

* Add three modes for prelu_op.