* add auto tuner

* fix prune

* fix sharding prune and mbs candidates

* fix cfg

* fix launch

* fix launch

* add unittest

* fix code style

* Fix bug of reduce_sum op. When input.numel() > INT32_MAX, its result
is wrong.

* Remove climits.

* Fix problem of pickle and NCCL_P2P_DISABLE in distributed testcases in
cuda12.

* Fix problem of TimeOut of distributed testcases under cuda12.

* Remove useless modification.

* Remove useless modification.

* move batch_norm prim test to op_test

* fix optest bug

* add test to cmake

* add cinn test case

* fix batch_norm prim grad bf16

* fix code

* add cuda check

* fix batch_norm bfloat16

* fix cpu bfloat16 bug

* skip non-bfloat16-supported platform

* fix code

* fix cinn rtol and atol in bfloat16

* fix name

* fix config

* [Zero-Dim] paddle.nanmedian support 0D Tensor

* fix CI

* update

* update

[BugFix]: Fix ci test bugs in test_fuse_gemm_epilogue_pass.py and test_fused_gemm_epilogue_op.py (#54519)

* fix ci bugs in fused_linear

* fix code style

* support mutable attr is input for build

* add ut

* solve conflict

* fix A100 CUDA12 ut

* fix ci uts

* fix test_sync_batch_norm_op

* fix sync bn op ut again by separating 2 files

* fix codestyle ci

* combine other PRs

* fix codestyle

* fix codestyle ci

* [IR]Polish ProgramTranslator private member code style

* update blog

[IR&PASS] part 3-2: add PatternApplicator and FrozenRewritePatternSet, refine PatternMatch code, add some api for Builder (#54492)

* [IR&PASS] add PatternApplicator and FrozenRewritePatternSet, refine PatternMatch code, add some api for Builder and TypeId

* fix comment

* “update”

* update

* update

* update

* update

* test

* update

* test

* fix_c++17_bug

* fix coverage compile error

* test

* test

* test

* fix C++17 error

* fix c++17 error

* fix c++17 error

* test

* test

* test

* test

* fix cinn compile error

* compile to compiler

* set cinn c++14

---------
Co-authored-by: huangjiyi <947613776@qq.com>
Co-authored-by: Nhuangjiyi <43315610+huangjiyi@users.noreply.github.com>

* construct dist tensor

* move constructor to header

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* Enable check_cinn on atan2, tile, top_k and where

* Update cmakelists in legacy_test

* Reformat code

* Enable check_cinn on op take_along_axis legacy test

* Enable check_cinn on pool2d

* Remove check_cinn=False

* Try fix tile test error

* Rename enable_cinn to test_cinn

* Refactor test_tile_op

* Replace all enable_cinn to check_cinn

* Revert pool2d test timeout

* Remove check_prim and use enable_cinn

* fix a100 cuda12 timeout

* fix cuda12 pickle loads problem

* fix ist_sharding_save ut

* mutiply_triple delete

* add case

* add timeout

This reverts commit 8c74ffc0.

* forbid fp64 enter into trt

* use global group to pass meta

* use batch isend irecv

* add partial send/recv

* remove communication group

* remove p2p on npu and xpu

* remove virtual pp ut

* add kernel dialect

* change DenseTensorTypeStorage to DenseTensorType

* add test case`

* add first pd_op to kernel dialect

* lower pd op to kernel dialect

* update

* update

* remove useless code

* add attrite print test

* fix bug

* polish code

* layer_norm op with dynamic shape support INormalizationLayer in TRT8.6

* Using trt layer to make layers_norm op in lower than trt8.6
layer_norm op with dynamic shape support INormalizationLayer in TRT8.6