* add gelu

* fix gelu

* add log_softmax

* add prelu kernel and prelu/gelu/logsoftmax infershape

* fix

* fix

* fix

* fix

* fix ci

* log_softmax rewrite

* fix

* fix

* fix conflict

* fix compile error

* fix comment

* fix

* ci_fix
Co-authored-by: NYan Li <liyan665@gmail.com>

* More precise mkldnn kernel choice in GetExpectedKernelType

* Fixes after review

* Refresh develop for CI

* CI experiment

* get back from CI exper

* Add bfloat16 softmax and gelu

* Add pass attr bfloat16_enabled_op_types

* Changes from review

* Delete is_test from activation operators
test=develop

* Revent unneeded changes
test=develop

add approximation for gelu, default value is False (only kernel with eigen is added, remove code for computing gelu with MKLDNN temporarily)

* refine grad maker, test=develop

* refactor tracer stage 1, test=develop

* merge develop to solve conflict third times, test=develop

* add erf op and python interface.

* add fp16 support for erf op.

* add unitests for erf op and its python interface.