Fix fused gemm bugs

[inference][cherrypick] Implement layer_norm op using INormalization Layer and conv_fusion support bias's rank equal to input's rank  (#54590)

* [inference]conv_fusion support bias's rank equal to input's rank (#54477)

* support bias's rank equal to input's rank

* [inference][trt]layer_norm op with dynamic shape support INormalizationLayer in TRT8.6 (#54379)

* layer_norm op with dynamic shape support INormalizationLayer in TRT8.6

* Using trt layer to make layers_norm op in lower than trt8.6
layer_norm op with dynamic shape support INormalizationLayer in TRT8.6

---------
Co-authored-by: Nbukejiyu <52310069+bukejiyu@users.noreply.github.com>

* [Distributed] Add pipeline opt memory (#54505)

* fix bug of release output in pp

* refactor scalar attribute

* fix op build

* fix merge conflicts

* fix coverage ci

* update roll convert

* [0D-Tensor] Support unary op, fix test_activation_op

* resolve conflict

* restore TestGelu

* polish codes according to comments

* sharding pp overlap

* bug fix

* update

* rename function

* update code logic

* add constant op

* support mutable attribute

* refine code

* fix bug

* fix bug

* refine code

* fix bug

* refine code

* refine code

* add ut

* refine code

* fix test bug

* solve conflict

* refine code

Support code generation for op conv2d_transpose, conv3d_transpose, depthwise_conv2d_transpose (#54242)

* update

* update eager_gen

* update

* rm intermediate

* adjust some API name

* fix inference.a undefined reference to utf8proc

* add vector type support for program translator

* polish

* support basic attribute type

* resolve conflicts

* add verify for combine/slice and unittests

* polish

* support more type in attribute translator

* modify by reviews

* fix merge mistakes

* refine code

* refine code

* add interface

* fix: op name normalization

* fix typo

* refactor input translator

* fix merge conflicts

* fix op normalizer bug

* refactor attribute translator

* fix bug

* refactor output translator

* fix typo

* fix

* fix approval error

* fix coverage

* fix op_compat parser

* fix merge conflicts

* fix merge conflicts

* fix merge conflicts

* fix merge conflicts

* fix merge conflicts

---------
Co-authored-by: Nzhangbo9674 <zhangbo54@baidu.com>

* remove reduce_sum_op.h

* support auto generate for static op reduce_sum

* remove reduce_sum_op in CMakeLists.txt

* add vector type support for program translator

* polish

* support basic attribute type

* resolve conflicts

* add verify for combine/slice and unittests

* polish

* support more type in attribute translator

* modify by reviews

* fix merge mistakes

* refine code

* refine code

* add interface

* fix: op name normalization

* fix typo

* refactor input translator

* fix merge conflicts

* fix op normalizer bug

* refactor attribute translator

* fix bug

* refactor output translator

* fix typo

* fix

* fix approval error

* fix coverage

* fix op_compat parser

* fix merge conflicts

* fix merge conflicts

* fix merge conflicts

* fix merge conflicts

* fix merge conflicts

* revert some changes

---------
Co-authored-by: Nzhangbo9674 <zhangbo54@baidu.com>

* 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

* [CustomDevice] add barrier op

* update

* change pyobject_isinstance

* change pyobject_isinstance

* support master_grad for adam and momentum

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

* function

* remove print

* add some pystate