Merge pull request !26832 from zichun_ye/akg_json_build

Changed the callback function GetProcessorFromContext to GetTargetFromContext,
so that we can use it to filter the clusterable op list, added a GetProcessorByTarget into AkgKernelJsonGenerator.

Moved the function IsKeepBasicNode, GetValidOps, OpListFilter from graph_kernel_helper
to graph_kernel_utils. combined the GetValidOps and OpListFilter.

Decoupled the pass getitem_tuple from "optimizer/common/helper.h", by deleting the checking
of input size. cnode->input(i) also checks the input index.

fix typo from code check

fix namespace error

add CollectFusedJsonWithSingleKernel to generate json for customp[

fix typo

fix typo

drop changes in CreateOutputsJson

the function IsOneOfPrimitive and IsOneOfPrimitiveCNode is useful,
we can move them into anf.cc

only the GET functions are implemented now.
remove the calling of AnfAlgo's GET functions for node info from AkgKernelJsonGenerator.

And, bugfix in pass reorder_ops, which set attrs for the same prim::Cast primitive in different CNode.

Merge pull request !25475 from DeshiChen/1025_genjson

* move GetInputTensorValue from common_utils to json_generator
* get dtype size by `Number.nbits()` instead of `GetDtypeNbyte` map.
* manually get attr from anfnode, instead of `AnfAlgo::GetNodeAttr`
* replace `AnfAlgo::GetCNodePrimitive` with `GetCNodePrimitive` in anf.cc
* it's not used to judge `AnfAlgo::IsRealKernel` in inner function.

cleancode jobs:
* remove the `Clean` function from AkgKernelJsonGenerator
* delete the json key "id", to delete the mutex in AkgKernelJsonGenerator

changed the function to "TypeIdToString", and use the Type::ToString() function,
instead of TypeId-String map.

changed the DtypeToTypeId together, the original StringToType can be used.

added a new interface StringToTypeId.

Merge pull request !25082 from DeshiChen/1015_set_kernel_attr

it's unreasonable to change the node when generating kernel json.
instead, it should be set in a pass.

most of the operators in original akg_kernel_attrs_process are not longer used,
so we deleted them, leaving only the "Cast" and "MatMul/BatchMatMul".

Merge pull request !23984 from looop5/custom_commit

only Linux system is supported now.

change the default value of `ENABLE_AKG` to off, and controlled by option `-K`.
the `ENABLE_AKG` is auto enabled when `ENABLE_GPU` or `ENABLE_D` is on.
since now, we can use `ENABLE_AKG` to control the compilation of graphkernel
and akg codes.

fix usage description for option "-K", it should be "[-K on|off]".

LLVM is required by akg for cpu kernels, so AKG for cpu is default disabled now.

This reverts commit b077aa1cab.

Revert "[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset"

This reverts commit 4e6f7dc97d.

delete pass_registry_test.cc

comment  hiai_nlu_model_multi.pb related  line

Merge pull request !21412 from chengbin/master

Merge pull request !21653 from ZengZitao/gk_fix_waring_master

2. delete unused function

Merge pull request !21364 from chengbin/bug_fix

Merge pull request !20230 from looop5/workspace

deal with workspace on cuda

cards

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++.

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 !18729 from lishanni/for_regis_limit

Refactor the original "PassManager" class, and derive the "GraphKernelPassManager"
GraphKernel's ir files are dumped into a new sub-directory "graph_kernel" in the original "verbose_ir_files"

All GraphKernel's passes are divided into 3 levels, and controlled by the flag "opt_level" by default.
when the opt_level is greaterequal to the pass's level, this pass will run.
The default "opt_level" is 2 when GraphKernel is enabled.
Levels:
  1. Basic features, like cluster, splitter, and some preprocess, postprocess.
  2. All stable features, mainly includes the optimization passes.
  3. Experimental features, like stitch-fusion, parallel-fusion.

The two flags "enable_pass" and "disable_pass" are available in this commit.
User can manually enable some passes when it's disabled by "opt_level", or disable the enabled passes,
by specifying that pass in this format: "stage_id.pass_id" or "stage_name.pass_name", multiple passes are separated by comma(",")
the stage/pass index and stage/pass name can be found from the ir filename.
e.g.   "--enable_pass=cluster.graph_kernel_expander,1.1,1.2"

Others:
  1. the pass "tensor_promotion" is not useful, remove it.
  2. put the pass "InsertPadOps" before "ArithmeticSimplify".