* add rmsnorm kernel
* add static graph test
* fix round type
* use alignas to avoid msvc compile error
* remove redundant headerfile to avoid rocm compile error
* fix rocm compile not found cub
* Add document

* fix bug

* fix bug

* fix bug

* rename weight_only/llm.int8

* [NewIR]Refine IrPrinter and basic Concept Interface for const Object

* move ut of auto_parallel

* fix ut

* add stop_gradient attribute for defining op

* modify by reviews

* fix

* add inplace interface

* support inplace

* refine code

* fix bug

* fix bug

* refien code

* add reduction spmd rule for auto parallel

* fix the logic of handling partial

* fix code style

* fix the partial handling

* [IR&PASS] add conv + elementwise_add fuse pattern

* add conv2dAddPattern to pass

* remove the extend_optimizer_with_weight_decay function

* Update __init__.py

* fix bug

* fix bug

* add some basic functions

* add elementwise rule for auto parallel

* add unit test for elementwise rule

* fix the lib name in spmd rule test cmake file

* fix some bugs

* add unit tests for elementwise spmd rule in python

* bug fix

* delete cpp unit test for elementwise spmd rule (use python ut now)

* add cpp unit test for elementwise rule

* use concrete op name in unit test

* fix typo

* fix code style

* delete cpp unit test

* add more details in log

* remove some pybind interface in cinn to fix compile problem

* modify cmake

* fix cmake

* add log for build cinn whl

* fix ninja for cinn

* fix conflict

* add ir output check in OpTest

* add ir grad check in op test

* fix legacy name converter bug

* add more unittest

* fix

* fix warprnn op bug

* add whit list

* polish code

* polish code

---------
Co-authored-by: Nkangguangli <kangguangli@hotmail.com>

* skip flash attn ut on Hopper

* minor change

* add clip_grad_value_ api

* add test for ClipGradByValue

* typo fix

* refine and modify clip_grad_norm_

* no_grad

* clip_

* remove g=p.grad

* bug: AssertionError: When Variable is used as the condition of if/while , Variable can only contain one element.

* fix prim add fill_any_like bug

* polish code

Pcard-72384
paddle with cinn的情况下支持cinn的c++ ut。 统一了cinn和paddle的third_party的c++版本，cinn third party不使用C++17

* add ir output check in OpTest

* add ir grad check in op test

* add more unittest

* fix

* Refactor op test mul

* Fix test error for mul

* add local scope

* refine code

* refien code

* refine code

* support local scope for BuildFuncList

* fix bug

* add log

* fix bug

* polish code

* fix bug

* masked select forward support broadcast

* cpu forward and backward

* gpu support mask broadcast

* fix comment

* x support broadcast

* fix comment

* configure elementwise_pow op_version

* support auto generate for static op elementwise_pow

* pre-commit run