* decouple reduce_any_op.h and reduce_op.h from reduce_any_op.cc

* support auto generate for static op reduce_any

* fix inference static lib bugs

* add if for copy

* fix py3 bugs

* change phi to static for inference

* fix py3 bugs

* fix ci-inference

* fix inference bugs

* perfect code

* revert to paddle op when matrix_multiply cannot enter into paddle-trt 

* fix mac typeinfo bugs

* add file

* move code to cc

* fix compile bugs

* add fp&bf16 bernoulli

* add check_dtype & fix error

* fix rocm error

* [CustomDevice] add barrier op

* update

* change pyobject_isinstance

* change pyobject_isinstance

* floor div support float/double/bfloat16/float16

* add ut

* fix bug

* fix fft.ifftshift for floor_divide upgrade

* fix comment

* fix bugs

* fix bug

* [IR] delete_wrapper

* [IR] modify the ast_transformer core logic

* [fix] modify the test_origin_info

* Optimize perf of broadcast matmul

* support more dtype

* [AutoParallel] add 1F1B

* rm amp

* support master_grad for adam and momentum

Co-authored-by: zhangting_2017@163.com <zhangting2020>

* function

* remove print

* add some pystate

* static graph autogen code for shape op

* fix onednn

* fix onednn

* [IR]Polish code in builtin type declaration

* fix typo

* add build

* add build

* refine code

* refine code

* refine code

* refine code

* refine interface

* fix bug

* fix bug

* fix bug

* refine yaml

* fix dropout api and support bf16 for prim

* fix code style

* fix dropout test

* fix dropout p = 0 test

* revert sequence_enumerate_op

* restore sequence_enumerate_op

* test

* test

* support layer_norm prim op bf16 dtype

* polish code

* resolve conflict

* [ROCM] fix multihead_matmul

* skip bf16 uts

* update

* static graph autogen code for check_finite_and_unscale_ op

* bug fixed