enable 6 rt_detr_v2 cases on xpu (#37548)

* enable 6 rt_detr_v2 cases on xpu

Signed-off-by: YAO Matrix <matrix.yao@intel.com>

* fix style

Signed-off-by: YAO Matrix <matrix.yao@intel.com>

---------

Signed-off-by: YAO Matrix <matrix.yao@intel.com>
Co-authored-by: Yih-Dar <2521628+ydshieh@users.noreply.github.com>

tests/models/rt_detr_v2/test_modeling_rt_detr_v2.py

@@ -27,7 +27,13 @@ from transformers import (
     is_torch_available,
     is_vision_available,
 )
-from transformers.testing_utils import require_torch, require_torch_gpu, require_vision, slow, torch_device
+from transformers.testing_utils import (
+    require_torch,
+    require_torch_accelerator,
+    require_vision,
+    slow,
+    torch_device,
+)
 from transformers.utils import cached_property
 
 from ...test_configuration_common import ConfigTester
@@ -636,7 +642,7 @@ class RTDetrV2ModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase
         self.assertTrue(not failed_cases, message)
 
     @parameterized.expand(["float32", "float16", "bfloat16"])
-    @require_torch_gpu
+    @require_torch_accelerator
     @slow
     def test_inference_with_different_dtypes(self, torch_dtype_str):
         torch_dtype = {
@@ -658,7 +664,7 @@ class RTDetrV2ModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase
                 _ = model(**self._prepare_for_class(inputs_dict, model_class))
 
     @parameterized.expand(["float32", "float16", "bfloat16"])
-    @require_torch_gpu
+    @require_torch_accelerator
     @slow
     def test_inference_equivalence_for_static_and_dynamic_anchors(self, torch_dtype_str):
         torch_dtype = {