e2e weak_object_expired GPU crash

### Describe the bug

Seen here: https://github.com/intel/llvm/actions/runs/13810199638/job/38630801652?pr=17189
```
-- Testing: 2284 tests, 10 workers --
FAIL: SYCL :: WeakObject/weak_object_expired.cpp (614 of 2284)
******************** TEST 'SYCL :: WeakObject/weak_object_expired.cpp' FAILED ********************
Exit Code: 134

Command Output (stderr):
--
RUN: at line 2: env ONEAPI_DEVICE_SELECTOR=hip:gpu /__w/llvm/llvm/build-e2e/WeakObject/Output/weak_object_expired.cpp.tmp.out
+ env ONEAPI_DEVICE_SELECTOR=hip:gpu /__w/llvm/llvm/build-e2e/WeakObject/Output/weak_object_expired.cpp.tmp.out
Memory access fault by GPU node-1 (Agent handle: 0x35916970) on address 0x763cb9d22000. Reason: Page not present or supervisor privilege.
/__w/llvm/llvm/build-e2e/WeakObject/Output/weak_object_expired.cpp.script: line 1: 2427 Aborted (core dumped) env ONEAPI_DEVICE_SELECTOR=hip:gpu /__w/llvm/llvm/build-e2e/WeakObject/Output/weak_object_expired.cpp.tmp.out

--
```
There's no reasonable way that's caused by the changes in https://github.com/intel/llvm/actions/runs/13810199638/job/38630801652?pr=17189 so I'm assuming this is flaky