* Support MaskedSelectGrad op with Kernel Primitive API

* Replace PADDLE_WITH_XPU2 with PADDLE_WITH_KP

* Add function description for Kernel Primitive API
1. Set cumsum and sort share memory size = 1024
2.sort and cumsum api limitation : blockDim.x must be less than 512 (blockDim.x <= 512)

* Modified RandomKernel with Kernel Primitive API

* update pten.h to phi.h

* update

* update fullKernel

* rename pten dir to phi

* rename namespace to phi

* rename infrt pten dir to phi

* resolve conflict

* rename pten to phi in cmake

* revert all infrt change

* change needed files

* fix infrt failed

* fix inference failed

* Unify paddle/pten::framework::ddim into pten::ddim

* fix paddle namespace

* compile sucessfully

* fix npu src file

* fix conflict

* fix conflict

* fix tensorrt compiler error

* fix conflict

* fix conflict

* fix tesst file conflict

* fix conflict

* fix mlu file conflict

* fix mlu file conflict

* fix cinn header file conflict

* fix conflict

* fix conflict

* fix conflict

* fix conflict

* improve backward performance

* support different dtypes for elementwise ops

* move kernel_primitives

* use pten's errors

* Migrate Dim and DDim from paddle::framework into pten namespace

* fix paddle::framework::Array

* fix framework::Array

* No harm to KP

* Pass the compile stage

* change the WriteData function

* fix template bugs and pass ctest of current elementwise

* for passing partial template specialization of tempalte function in CI-ROCm

* To make 'WriteData' funtion flexible.

* a less harmful way to support multi-output

* a less harmful way to support multi-output

* Modify reduce_op.op.h for xpu2 with kernel primitive api

* Update the implement of reduceAnyKernel according to kernel primitive api
* Fix a bug in ReadData, ReadDataBc and ReadDataReduce when NX != 1

* add ElementwiseTernary, Reduce, ReadDataStride

添加Kernel primitives api： ReadData, WriteData ComputeFunctor

This PR add some primitives api related files. Those files are almost empty for the moment, but will be filled later.

* Add dlnne engine runtime

* Fix log

* Remove <const_cast> and remove unrelated modify with dlnne, +clang-format

* Fix CMakeList format error

* Add copyright message

* Fix dlnne CMakeList.txt

* Add some paddlepaddle_pass to support more networks

* Fix some format bug

test=develop

add python inference api

fix cmake again
test=develop

Add the demo of subgraph splitter

* Update tensor_util.h

* Update with moved TensorDesc

* Fix tensur_utils.cu

* Update

* Update

* Update

* Update

* Make tensor_util.cu a symbolic link