* [Sparse] add SparseCsrTensor fused_attention kernel and API

* fix comment

* [Sparse]add SparseTensor mv kernel(csr*dense_vec->dence_vec, coo*dense_vec->dense_vec)

* fix CI

* perfect copy

* deal with conflict

* deal with conflict

* fix compile bugs

* fix unittest bugs

* change code format

* deal with conflict

* modify code by review

* fix ce bugs

* fix ce bugs

* add lo

* perfect code format

* deal with conflicts

resort .cu headers, set clang-format not sort include block and consider .cu as main source file (#43633)

* add new API/OP(csr->csr) of SparseTensor softmax

* fix comment

* fix to_sparse_csr

* add some kernel(csr*dense->csr, dense*dense->csr) of SparseTensor matmul

* fix CI

* fix CI

* fix comment

* fix comment

* use GpuMemcpy and GpuMemset

* sparse convert kernel support double dispatch by indices dtype

* cudaMemcpyKind->gpuMemcpyKind

* remove storage

* add glog include

* add glog include

* add glog include

* remove shared_storage

* fix bug

* fix rnn bug

* chnage dispatch to visit

* resolve conflict

sparse maxpool; kernel_registry support sparse tensor

restructure conv

* move the common function to kernel/funcs/sparse/

* add namespace

* rm unused file

* move func

* reuse code

submanifold convolution

Sparse conv3d backward(gpu)

* sparse conv3d: gpu code

* fix incorrect dims settings

* sparse conv3d

* fix out dims

* test performance

* test large shape success

* opt scatter, double performance

* test float16

* remove profiling code

* remove pten

* opt code lines

* correct boundary judgment

* only cpu

* test ci

* test ci

* remove the including paddle/fluid header; extract the conmmon function

* opt code lines

* use DenseTensor::data() instead of mutable_data

* return rulebook for backward

* specify layout

* rename:conv -> sparse_conv3d

* change Vector to std::vector and provide MixVector class as a helper wrapper class

* solve the multi-gpu hang problem

* remove the duplicate template instantialize

* Copy vector to cpu

* add CopyToCPU

* xxx

* final version: fix the problem of all reduce

* remove mixvector dependence

* fix

* merge

* fix code

* fix by CI

* unify register macro

* rename declare macro

* fix infrt error

* 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