crash still occur in sdpa

### 🐛 Describe the bug

see in tests/pipelines/qwenimage/test_qwenimage_edit.py::QwenImageEditPipelineFastTests::test_num_images_per_prompt
```
import torch

batch_size, seq_len, num_heads, head_dim = 4, 16413, 3, 16
torch.manual_seed(0)

query = torch.randn(batch_size, seq_len, num_heads, head_dim, device="xpu", dtype=torch.float32)
key = torch.randn(batch_size, seq_len, num_heads, head_dim, device="xpu", dtype=torch.float32)
value = torch.randn(batch_size, seq_len, num_heads, head_dim, device="xpu", dtype=torch.float32)

# torch SDPA expects [batch, heads, seq, head_dim] for this usage.
query_t = query.permute(0, 2, 1, 3)
key_t = key.permute(0, 2, 1, 3)
value_t = value.permute(0, 2, 1, 3)

out = torch.nn.functional.scaled_dot_product_attention(
query=query_t,
key=key_t,
value=value_t,
attn_mask=None,
dropout_p=0.0,
is_causal=False,
scale=None,
enable_gqa=False,
)

# Convert back to [batch, seq, heads, head_dim] to match the original layout.
out = out.permute(0, 2, 1, 3)
torch.xpu.synchronize()
```

Segmentation fault from GPU at 0xff00000000094000, ctx_id: 1 (CCS) type: 0 (NotPresent), level: 1 (PDE), access: 1 (Write), banned: 1, aborting.
Segmentation fault from GPU at 0xff00000000094000, ctx_id: 1 (CCS) type: 0 (NotPresent), level: 1 (PDE), access: 1 (Write), banned: 1, aborting.
Abort was called at 288 line in file:
./shared/source/os_interface/linux/drm_neo.cpp

### Versions

Versions of relevant libraries:
[pip3] apollo-torch==1.0.3
[pip3] dpcpp-cpp-rt==2025.3.2
[pip3] galore-torch==1.0
[pip3] impi-rt==2021.17.2
[pip3] intel-cmplr-lib-rt==2025.3.2
[pip3] intel-cmplr-lib-ur==2025.3.2
[pip3] intel-cmplr-lic-rt==2025.3.2
[pip3] intel-opencl-rt==2025.3.2
[pip3] intel-openmp==2025.3.2
[pip3] intel-pti==0.16.0
[pip3] intel-sycl-rt==2025.3.2
[pip3] mkl==2025.3.1
[pip3] mypy_extensions==1.1.0
[pip3] numpy==2.2.6
[pip3] oneccl==2021.17.2
[pip3] oneccl-devel==2021.17.2
[pip3] onemkl-license==2025.3.1
[pip3] onemkl-sycl-blas==2025.3.1
[pip3] onemkl-sycl-dft==2025.3.1
[pip3] onemkl-sycl-lapack==2025.3.1
[pip3] onemkl-sycl-rng==2025.3.1
[pip3] onemkl-sycl-sparse==2025.3.1
[pip3] onnx==1.21.0
[pip3] pytorch-msssim==1.0.0
[pip3] tbb==2022.3.1
[pip3] tcmlib==1.4.1
[pip3] torch==2.13.0.dev20260419+xpu
[pip3] torchao==0.18.0.dev20260419+xpu
[pip3] torchaudio==2.11.0.dev20260419+xpu
[pip3] torchcodec==0.11.0+cpu
[pip3] torchdata==0.11.0
[pip3] torchsde==0.2.6
[pip3] torchvision==0.27.0.dev20260419+xpu
[pip3] triton==3.6.0
[pip3] triton-xpu==3.7.1+git21033c4e
[pip3] umf==1.0.3