* Implement conv2d_fusion NHWC format using CUTLASS
* Add unit testing for CUTLASS Conv in inference
* Add experimental API for CUTLASS.

* [OpAttr]Fix Ignore AttriteTensor in IndicateDataType bug in grad_op

* add GetExpectedKernelType

* Use BroadcastKernel and ReduceKernel to optimize expand and expand_grad.

* Correct the axis when there is only 1 input in BroadcastKernel.

* Add the calculate of output's shape.

* [Zero-Dim] reshape/reshape_/reverse 0D support

* rm comment

* change paddle.to_tensor to paddle.full

* fix docs

* update paddle.full

* relu flops all

* add annotations and tests

* revision for codestyle

* bugfix: fix bugs in Indexable and support LayerDict

* fix bugs.

* check weight shape of conv1d_transpose

* add unittest case

* speedup getFNDAFile

* add fnda_base for c++ ut cc file

* fix bug

* fix bug

* fix bug

* fix bug

* add custom_cpu testcase

* update test_custom_device_setup

* update path to custom_runtime

* fix cmd wait

* test Linux only

* setup once

* integrate to one run_cmd

* add pip install

* change timeout

* add debug string

* add debug string

* add debug string

* use os.system and change module name

* add runtime

* add more debug message

* continue debug

* timestamp

* fix testcase import bug

* remove error message

* set TIMEOUT property

* fix test_conv_bn_fuse_pass_cc

* remove comment

* update conv to convNd

* trigger ci

* delete batch_norm

* test

* test

* test

* test

* test

* recover cmake_gen

* debug

* support static graph code-gen for squeeze op

* generate static graph code of unsqueeze

* refine op name

* add extra output in op_compat

* remove debug log

* clean custom_xpu testcase test_static_pe

* use assert_allclose to solve precision error

* adjust precision

* flatten tensor

* fix flatten

* 1219

* temporarily change the num_diff_files limit, test=document_fix

* Revert "temporarily change the num_diff_files limit, test=document_fix"

This reverts commit 8e70f00ef468d2dad0e38b3da06295ed62990d20.

* for codestyle

* remove duplicate license

* `static mode` -> `static graph mode`

* Update hybrid_parallel_inference.py

* Update layer_function_generator.py

* Update manipulation.py

* reset
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>
Co-authored-by: NSigureMo <sigure.qaq@gmail.com>

* fix_mac_build_problem

* fix_mac_build_problem

* fix_mac_build_problem

* Fix default GetExpectedKernelType for ops supported tensor attrs

* rm legacy

* clear in_legacy

* fix tracer

* fix_static_problem

* test

* fix_static_problem,test=document_fix

* [fluid remove] rawconv

* [D2SCinn]Support deliver skip_gc_vars into Graph

* fix unittest

* fix copy

* auto parallel bf16

* fix load into memory

* fix load into memory

* fix code style