[HAL/AMDGPU] Use status matcher in notification test

Use the IREE status test macro for notification-ring callback status validation instead of checking the raw boolean status predicate. This keeps AMDGPU tests aligned with the runtime test style and preserves the diagnostic status payload on failure.

runtime/src/iree/hal/drivers/amdgpu/util/notification_ring_test.cc

diff --git a/runtime/src/iree/hal/drivers/amdgpu/util/notification_ring_test.cc b/runtime/src/iree/hal/drivers/amdgpu/util/notification_ring_test.cc
index 25f5701..6579861 100644
--- a/runtime/src/iree/hal/drivers/amdgpu/util/notification_ring_test.cc
+++ b/runtime/src/iree/hal/drivers/amdgpu/util/notification_ring_test.cc
@@ -48,7 +48,7 @@
 static void VerifySemaphoreNotVisibleBeforePreSignalAction(
     iree_hal_amdgpu_reclaim_entry_t* entry, void* user_data,
     iree_status_t status) {
-  EXPECT_TRUE(iree_status_is_ok(status));
+  IREE_EXPECT_OK(status);
   EXPECT_NE(entry, nullptr);
   auto* state = static_cast<PreSignalActionState*>(user_data);
   EXPECT_EQ(iree_async_semaphore_query(state->semaphore), 0u);