[ci]: blocklist test_autograd for core

[yuzibo]

🐛 Describe the bug

On v100, ninja will crash when running ninja:

home/jenkins/pip/bin/python3
/home/jenkins/pip/bin/ninja
Python 3.13.5
/home/jenkins/scripts/shard_with_label.sh: line 128:  1643 Segmentation fault      ninja --version

see https://community-ci.openruyi.cn/ruyici/job/PytorchCi/job/10-k1-shard-ci/2941/console

this will lead the case failed:

"test_autograd.py::TestAutograd::test_multi_grad_all_hooks"

see log

Unblock the process to upstream, so add it temporarily.

And it is okay to run with the same CI job. Cannot reproduce it on v100 manually also.

Versions

================ collect_env.py ================
Collecting environment information...
PyTorch version: 2.15.0a0+gited5a91a
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: Debian GNU/Linux 13 (trixie) (riscv64)
GCC version: (Debian 14.2.0-19) 14.2.0
Clang version: 19.1.7 (3+b1)
CMake version: version 3.31.6
Libc version: glibc-2.41

Python version: 3.13.5 (main, Jun 13 2026, 14:18:01) [GCC 14.2.0] (64-bit runtime)
Python platform: Linux-7.0.0-29-generic-riscv64-with-glibc2.41
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
Is XPU available: False
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: False
Caching allocator config: N/A

[yuzibo]

This issue is reproducible on v100 and k1