Support return_tensors in audio chat templates (#34601)

* add audio chat templates

* update

* update

* nit

* green ci

* we dont care about the order anymore

* clean up after rebase

* overriden tests rename

* rename shieldgemma also

* one more rename

* require_read_token

* removde images/videos

* retrigger CI flaky

tests/models/aria/test_processor_aria.py

@@ -238,7 +238,7 @@ And who is that?<|im_end|>
         self.assertEqual(rendered, expected_rendered)
 
     # Override as AriaImageProcessor doesn't accept `do_rescale`
-    def test_chat_template_accepts_processing_kwargs(self):
+    def test_image_chat_template_accepts_processing_kwargs(self):
         processor = self.get_processor()
         if processor.chat_template is None:
             self.skipTest("Processor has no chat template")