Port pybind11 and python code to support py3 CI test

Add libpng dependencies to yum and correct libnccl dir

Refine gru cpu forward

Error message refine: add demangle api to attribute type

polish high frequency enforce error message

* remove patchelf in windowls

* "follow comment"

fix specific cudnn include and library path

* Remove Data Sharing between input and output in scatter_op

* Removed data sharing in backward op

* "cherry picked operators changes"

* "remove duplicated code"

* "add constant setter"

* "add get expected kernel"

* "fix ci"

* "add fill constant"

* Add three modes for prelu_op.

refine fc op

lookup table size fix

fix inference doc url error

* dist transpile can pass startup program by argument

* update API.spec

Add flatten op interface and enhance APIs about detection to support variable-length image. (#12422)

* add flatten api&enhance detection api

* unify shape_op data type

* update API.spec

Make cross_entropy_op supporting tensor

Polish API of exception holder

test and doc IR Graph