[CI] Fix broken CI (#2302)

1. disable test_eagle_ccorrectness test, we'll reopen it once oom error
fixed.
2. drop transformers version limit for main, since vLLM rely on
>=4.55.0, see:
65552b476b
3. fix kv_connector_output bug, see:
796bae07c5

- vLLM version: v0.10.0
- vLLM main:
d1af8b7be9

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>

vllm_ascend/worker/model_runner_v1.py

@@ -1605,9 +1605,12 @@ class NPUModelRunner(LoRAModelRunnerMixin):
                                                      intermediate_tensors))
         kv_connector_output = None
         if not vllm_version_is("0.10.0"):
-            kv_connector_output = KVConnectorOutput(
-                finished_sending=finished_sending,
-                finished_recving=finished_recving)
+            if finished_sending is not None and finished_recving is not None:
+                kv_connector_output = KVConnectorOutput(
+                    finished_sending=finished_sending,
+                    finished_recving=finished_recving)
+            else:
+                kv_connector_output = None
             finished_sending = None
             finished_recving = None
         with ProfileExecuteDuration().capture_async("post process"):