Merge pull request !21208 from baihuawei/graph_mode_nonsink_part2

Merge pull request !16452 from StyleHang/BandpassBiquadOp

Merge pull request !20995 from chenweifeng/cpu-dynamic-input

Merge pull request !16458 from StyleHang/AllpassBiquadOp

Merge pull request !21036 from yanghaoran/upgrade_ascend_0728

Signed-off-by: zhoufeng <zhoufeng54@huawei.com>

Merge pull request !16449 from StyleHang/BandBiquadOp

Merge pull request !20812 from ZPaC/cross-silo

Merge pull request !20672 from lianliguang/add_dump_func

Merge pull request !18674 from zetongzhao/deserialize

Add a subdirectory "model" in the "backend/optimizer/graph_kernel" for litegraph.
Implement two interfaces "AnfGraph2LiteGraph" and "LiteGraph2AnfGraph".

The litegraph will be the base data structure when we migrate the GraphKernel code
from python("mindspore/_extends/graph_kernel") to c++.

Merge pull request !20174 from DeshiChen/0708_compilewithjson

1. Add a new message type "AKG/ATTR" in AkgKernelBuilder.
   the attrs was sent before the kernel infos.
2. Send "online_tuning" attribute when the flag is not zero,
   but error occurs in the latest akg submodule.
3. Send "repository_path" attribute when the flag is not empty.
4. Add a new value "compute_capability" into kernel info when the processor is GPU.

Merge pull request !20246 from tanghuikang/error_manager

Merge pull request !19904 from laiyongqiang/fusionType

Merge pull request !19882 from yeyunpeng2020/new_api

Merge pull request !19764 from harshvardhangupta/rem-build-2

Merge pull request !19936 from VectorSL/signal

Merge pull request !19748 from xiaotianci/fix_rotate_center

Merge pull request !19771 from ehaleva/API2

Merge pull request !19507 from lixiachen/remove_cacheop_builders

Merge pull request !19563 from yyuse/fl_op_armour_dir

Merge pull request !18987 from lvchangquan/transdata_formal

Merge pull request !19458 from lichen/fix_pipeline_split_bug

Merge pull request !19047 from harshvardhangupta/remove_builder

Merge pull request !19457 from ZPaC/change-fl-dir