功能：本PR的目标是提高attention模块的计算性能。
为了减少框架层对op的调度开销，本PR通过在C++层手动实现attention模块，对外提供attention 大op；
为了减少防存开销，本PR采取了两种优化方法：
（1）在q,k,v计算时通过共享输入X，将该处的gemm，transpose和bias add从三次调用减少为一次；
（2）使用kernel融合优化技术，在不同cuda kernel之间通过寄存器传输数据；

* [hapi] support dygrapg amp O2

* fix problem of static pure fp16 in hapi

* fix bug

* fix format

* fix ut

* follow comments

* update ut

* update amp save/load

* fix ut

* refine code format

* Support elementwise_add triple grad Kernel

* Change code-format to follow CI std

* Removed unreasonable code, and fixed an input uninitialized issue

* Support elementwise_add triple grad Kernel

* Change code-format to follow CI std

* Removed unreasonable code, and fixed an input uninitialized issue

* added base changes for matmul_v2+trans+resh fuse pass

* added full matmul_v2+transpose+reshape pass

* removed a file added by mistake

* added reviewers suggestions

* Changed ops type in checking capatibility version

* Deteled one statement

* add sync_batch_norm (support train, infer, and fp32, fp16, and NCHW, NHWC)

* [NPU] Delete debug codes

* [NPU] Remove FP16

* add viterbi decode cpu kernel

* add viterbi decoder api in paddle.text

* add a data buffer once to avoid create many small pieces of data buffer frequently

* fix viterbi max_seq_length bug

* fix seq_len=1 bug

* fix device context

* move split out of for loop

* remove INVERSE_SUB

* remove 2 GET_CAST_MASK

* remove 1 loop

* remove Functor

* add to_static deploy code

* use MAX_FUNC instead of ELE_MAX

* add MaxFunctor

* impl max_func

* remove MaxFunctor

* remove cast op

* use REGISTER_OP_WITHOUT_GRADIENT

* add viterbi cuda kernel

* add FIX_BLOCKDIM_CASE macro

* add MKL add, mul; add get data mask

* add arange mkl impl

* add CPU Argmax

* add cpu gather

* use EXECUTE_MKL_ELEMENT_BINARY_OP instead of some ADD, MUL

* use SameDimsBinaryOP instead of EXECUTE_MKL_ELEMENT_BINARY_OP

* use SAME_DIMS_ELEMENT_BINARY_OP

* add SimpleBroadcastBinaryOP

* use int instead of int64_t to accelerate

* optimize SimpleBroadcastBinaryOP

* optimize SimpleBroadcastBinaryOP

* optimize performance in both single thread and multithread situation

* remove useless line

* remove useless code

* add CREATE_TENSOR_BUFFER macro

* add INIT_REQUIRED_TENSOR macro

* add comment

* fix windows ci

* add viterbi unittest

* remove cuda add functor

* remove cuda equal

* remove a template function

* fix windows ci

* fix windows dtype

* remove some template instance

* remove useless header file

* remove some blockdim

* remove transpose impl

* accelerate cpu performance on single thread situation

* viterbi_decode->crf_decode

* rename crf params name

* add viterbi api test

* remove useless import

* add enable_static

* use viterbi decoder

* fix viterbi len=1

* fix  viterbi unittest

* remove useless comments

* reconstruct viterbi decode

* remove ADD,SUB,MUL structure

* fix coverage

* remove CREATE_TENSOR

* add name args

* crf.py->ops.py; with_start_stop_tag->include_start_end_tag

* update crf_decode en docs

* fix viterbi decode en docs

* fix some review comments

* add FIXED_BLOCK_DIM_CASE in cuda

* push_back->emplace_back

* crf_decode->viterbi_decode; include_start_end_tag->include_bos_eos_tag

* paddle.text.ops.viterbi_decode->paddle.text.viterbi_decode

* fix viterbi_decode en docs

* add some ops to train ssd on kunlun

* update test_fill_any_like_op_xpu.py

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

* add ctr table depends

* code style

* fix

* fix

* fix naming

* rename

* rename

* add align for WorkQueue

* add spinlock

* merge develop

* merge

* Add EventsWaiter

* Revert "Add EventsWaiter"

This reverts commit e206173aa9be7401b83a53581627bfaf557c8fb2.

* adjust multithread using, fix flame graph

* update

* Add kQueueSync.synchronize_run_ logic

* Support No DataTransform From GetKernelTypeForVar

* split into PreBuildTask and BuildPull; slove endpass bug;test=develop

* change buildcpu into prebuild and buildcpu into build;test=develop

* fix global gather and global scatter operators

Add Tokenizer related functionalities for Transformer model in order that the process of training and predicting is consistent.

* support the text string as an input Tensor
* support the "VOCAB"unordered_map<wstring, int> as an input Tensor to lookup tokens
* Tokenizer used for BERT. This tokenizer applies an end-to-end, text string to wordpiece tokenization.
* It first applies basic tokenization, followed by wordpiece tokenization.

* bug fix for  DeserializeSelectedRows. test=develop

* fix bug for SerializeSelectedRows. test=develop

* update. test=develop

Add CINN compile option in CMake.

Now you can use CINN in Paddle by `-DWITH_CINN=ON` when `cmake`

To test it, you can run `make cinn_lib_test -j` and `ctest -R cinn_lib_test`. 

Note:
1. You should set
```
export runtime_include_dir=${CINN_SOURCE_DIR}/cinn/runtime/cuda 
```
When run test, the `${CINN_SOURCE_DIR}` should be set based on your CINN directory.

2. CINN is under developing now, you may have to change `CINN_GIT_TAG` to the git commit you need.

* fix

* remove const

* Support elementwise_add triple grad Kernel

* Change code-format to follow CI std

* [NPU] Add iou_similarity op

* [NPU] Add iou_similarity op

* [NPU] Add iou_similarity op

* [NPU] update inference cmake, test=develop

* address review comments, test=develop

* fix compile error when WITH_ASCEND_CXX11 ON, test=develop

* update

* fix ut error

* update ut

* add feed op and new var for the generated subgraph

* perfect the test script of build_cinn_pass 

* remove useless clear and perfect some annotation

* add nearest_interp_v2 trt plugin

* fix replicate pad when input size is 0
* add unit test

* Add QR decomposition op

* Change codes to adapt to new svd_helper

* Update linalg.py

Restore the deleted comma

* Restore the deleted line

* Update linalg.py

* Update linalg.py

* Improve the qr code by reviews

* Update QR based on CI results

* Update qr doc, test=document_fix

* Change unsafe and ill-formed codes