[HF][fix] Use ParameterizedModelParser instead of ModelParser

This seems like a pretty brittle bug. If we use `ModelParser` isntead of `ParameterizedModelParser`, we get a bug because `callback_manager` is not a supported kwargs for the `run_inference()` method. This gets filtered out by the `ParameterizedModelParser.run()` method (https://github.com/lastmile-ai/aiconfig/blob/0ceb17636bae2b5416e7b415cdc87fd71b5ba3b0/python/src/aiconfig/default_parsers/parameterized_model_parser.py#L48-L54) but these kwags are not accepted  by the `ModelParser.run()` method (https://github.com/lastmile-ai/aiconfig/blob/0ceb17636bae2b5416e7b415cdc87fd71b5ba3b0/python/src/aiconfig/model_parser.py#L62-L69)

## Test Plan

Before

After

extensions/HuggingFace/python/src/aiconfig_extension_hugging_face/local_inference/automatic_speech_recognition.py

@@ -3,15 +3,17 @@
 import torch
 from transformers import pipeline, Pipeline
 from aiconfig_extension_hugging_face.local_inference.util import get_hf_model
-from aiconfig import ModelParser, InferenceOptions
+
 from aiconfig.callback import CallbackEvent
+from aiconfig.default_parsers.parameterized_model_parser import ParameterizedModelParser
+from aiconfig.model_parser import InferenceOptions
 from aiconfig.schema import Prompt, Output, ExecuteResult, Attachment
 
 if TYPE_CHECKING:
     from aiconfig import AIConfigRuntime
 
 
-class HuggingFaceAutomaticSpeechRecognitionTransformer(ModelParser):
+class HuggingFaceAutomaticSpeechRecognitionTransformer(ParameterizedModelParser):
     """
     Model Parser for HuggingFace ASR (Automatic Speech Recognition) models.
     """
@@ -85,7 +87,7 @@ async def deserialize(
         await aiconfig.callback_manager.run_callbacks(CallbackEvent("on_deserialize_complete", __name__, {"output": completion_data}))
         return completion_data
 
-    async def run(self, prompt: Prompt, aiconfig: "AIConfigRuntime", options: InferenceOptions, parameters: Dict[str, Any]) -> list[Output]:
+    async def run_inference(self, prompt: Prompt, aiconfig: "AIConfigRuntime", options: InferenceOptions, parameters: Dict[str, Any]) -> List[Output]:
         await aiconfig.callback_manager.run_callbacks(
             CallbackEvent(
                 "on_run_start",

extensions/HuggingFace/python/src/aiconfig_extension_hugging_face/local_inference/image_2_text.py

@@ -10,8 +10,9 @@
 
 from aiconfig_extension_hugging_face.local_inference.util import get_hf_model
 
-from aiconfig import ModelParser, InferenceOptions
 from aiconfig.callback import CallbackEvent
+from aiconfig.default_parsers.parameterized_model_parser import ParameterizedModelParser
+from aiconfig.model_parser import InferenceOptions
 from aiconfig.schema import (
     Attachment,
     ExecuteResult,
@@ -24,7 +25,7 @@
     from aiconfig import AIConfigRuntime
 
 
-class HuggingFaceImage2TextTransformer(ModelParser):
+class HuggingFaceImage2TextTransformer(ParameterizedModelParser):
     def __init__(self):
         """
         Returns:
@@ -117,7 +118,7 @@ async def deserialize(
         await aiconfig.callback_manager.run_callbacks(CallbackEvent("on_deserialize_complete", __name__, {"output": completion_params}))
         return completion_params
 
-    async def run(self, prompt: Prompt, aiconfig: "AIConfigRuntime", options: InferenceOptions, parameters: Dict[str, Any]) -> list[Output]:
+    async def run_inference(self, prompt: Prompt, aiconfig: "AIConfigRuntime", options: InferenceOptions, parameters: Dict[str, Any]) -> List[Output]:
         await aiconfig.callback_manager.run_callbacks(
             CallbackEvent(
                 "on_run_start",