[V0 Deprecation] Remove V0 executors (#27142)

Signed-off-by: Nick Hill <nhill@redhat.com>

vllm/v1/engine/core.py

@@ -60,7 +60,7 @@ from vllm.v1.engine.utils import (
     EngineZmqAddresses,
     get_device_indices,
 )
-from vllm.v1.executor.abstract import Executor
+from vllm.v1.executor import Executor
 from vllm.v1.kv_cache_interface import KVCacheConfig
 from vllm.v1.metrics.stats import SchedulerStats
 from vllm.v1.outputs import ModelRunnerOutput
@@ -322,7 +322,6 @@ class EngineCore:
         with self.log_error_detail(scheduler_output):
             model_output = self.model_executor.execute_model(scheduler_output)
 
-        assert isinstance(model_output, ModelRunnerOutput)
         engine_core_outputs = self.scheduler.update_from_output(
             scheduler_output, model_output
         )
@@ -364,7 +363,7 @@ class EngineCore:
         if self.scheduler.has_requests():
             scheduler_output = self.scheduler.schedule()
             future = self.model_executor.execute_model(scheduler_output, non_block=True)
-            batch_queue.appendleft((future, scheduler_output))  # type: ignore[arg-type]
+            batch_queue.appendleft((future, scheduler_output))
 
             model_executed = scheduler_output.total_num_scheduled_tokens > 0
             if (
@@ -463,14 +462,6 @@ class EngineCore:
     ) -> list[_R]:
         return self.model_executor.collective_rpc(method, timeout, args, kwargs)
 
-    def save_tensorized_model(
-        self,
-        tensorizer_config,
-    ) -> None:
-        self.model_executor.save_tensorized_model(
-            tensorizer_config=tensorizer_config,
-        )
-
     def preprocess_add_request(self, request: EngineCoreRequest) -> tuple[Request, int]:
         """Preprocess the request.