* [CustomDevice] add pipeline parallel support

* update

* update

* delete overwrite from gather_grad

* fix

* Update gather_grad_kernel.cc

* remove axis from elementwise_grad

* Update elementwise_sig.cc

* relocate metri_op.py

* reloacte nn.py

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix variable->tensor and fix __all__

* fix ctr_metric_bundle and sparse_embedding

* fix bug of function init

* fix bug of importing sparse_embedding and ctr_metric_bundle

* fix bug

* Update __init__.py

* add tile_grad composite rule

* support auto generate for op adam optimizer

* remove unnecessary files

* support auto generate for op adamw optimizer

* support auto generate for op merged_adam optimizer

* use manual_signature in adam_

* remove ASCEND* keyword

* update docstring

* bug fixed

* bug fixed

* update

* update

* Revert "update"

* fix bug

* update

* trans fused attention to phi

* add optional parm

* trans fused_attention_grad to phi

* add fused attention grad register info

* fix include

* test=kunlun

* add fused attention to static build list

* add remove

* update remove

* unify_kernel

* fix compile bugs

* fix py3 bugs

* fix xpu bugs

* update

* fix bug

* fix bug

* fix bug

* fix bug

* first commit

* restruct c++ interface to divide linear from matmulwithcublaslt

* finish building in cublaslt impl

* fix code bugs

* fix host cost

* add some changes

* add autogen code support for mean_all op

* bug fixed

* bug fixed

* bug fixed

* fix graph_reindex

* fix

* Update op_compat.yaml

* update

* fix bug

* update

* fix bug

* update

* fix bug

* Add prelu, clip_by_norm, multi_dot tests

* Fix code

* Fix code

* add gn vjp

* fix 0

* fix args num

* fix type

* debug2

* remove unused expand

* support fp16

* fix typo

* fix reshape bug

* test3

* test4

* fix bug3

* add comment

* update

* fix bug

* update

* fix bug

* update

* revert lookup_table_op

* add autogen code support for rnn op

* fix bug

* fix bug

* add autogen code support for lu

* fix bug

* fix bug

* fix bug

* fix bug

* [CustomDevice] add c_identity op

* fix use calc stream

* unify kernel

* fix ci bugs

* fix py3 bugs

* fix py3 bugs

* perfect code

* fix bug in parse args

* fix bug

* recover legacy_*.yaml

* change 'Out' to Output

* [CINN] fix concat&pow

* update concat

* composite_backward_api

* for ci

* for ci

* update test & fix opmaker

* add register info for eigh and eig_gard

* add sync_batch_norm_op.cu register info

* add lamb output register info

* add unique register info

* change type name

* change type name

* add output register info for check_finite_and_unscale

* update cmake and config file

* add register info for adagrad

* fix build error

* add sync to run_unittests.sh

* add register info for unique_consecutive

* fix build error

* add eigh to STATIC_BUILD_TESTS

* update eig_kernel.cc

* update eig_kernel.cc

* fix infer mate error

* fix unique register error

* fix lamb register info error

* fix lamb register info

* update lamb register info

* fix lamb

* remove one Output Register

* update static build file

* add eigh op to disable_wingpu_test

* update run_unittests

* [AMP OP&Test] Sync_batch_norm support bfloat16

* fix

* fix

[Zero-Dim] support 0-D tensor for reduce/reshape/stack/prelu/expand_v2/gaussion onednn kernels (#52185)

* support 0-D tensor for reduce/reshape/stack/prelu/expand_v2/gaussion ops

* fix gaussian random mkldnn op ut

1. modify set_value op, use Scalars to represent attr `values`, instead of a bunch of attributs of various types; (#52408)

2. add program converter and set_value op as an example, which provides the functionality to convert `paddle::framework::ProgramDesc` between old and new formats(the differences are mainly some operators with incompatible updates in the definition);
3. program version and operator version map now are always saved when serializing `paddle::framework::ProgramDesc` to identify the version;
3. provide an option `legacy_format=false` in  serialization of `paddle::framework::ProgramDesc`, it decided whether to convert ProgramDesc back to a legacy format, which is compatible for paddle 2.4.2 or earlier versions to load and execute;
4. deserialization of `paddle::framework::ProgramDesc` is now automatically detecting whether the bytes it receives is in legacy format(contains any of the operators that has been incompatibly updated and have any attribute of type `Scalar`) and convert it to new format. But if you want a faithful deserialization without the automatic conversion, you can use protobuf's deserialization instead. Though it is not recommended, it can be used for the purpose of testing.

* [phi] move sequence_pool kernel to phi

* [phi] mv sequence_pooling to phi funcs

* [phi] mv sequence_pooling_test

* [phi] RollBACK `paddle/fluid/operators/sequence_ops/sequence_pool_op.cc`

* [phi][funcs] fix mutable_data

* [phi][funcs] fix mutable_data