chore: refactor (chat)completions endpoints to use shared params struct (#3761)

# What does this PR do?

Converts openai(_chat)_completions params to pydantic BaseModel to
reduce code duplication across all providers.

## Test Plan
CI









---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with
[ReviewStack](https://reviewstack.dev/llamastack/llama-stack/pull/3761).
* #3777
* __->__ #3761

llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/llm_as_judge_scoring_fn.py

@@ -6,7 +6,7 @@
 import re
 from typing import Any
 
-from llama_stack.apis.inference import Inference
+from llama_stack.apis.inference import Inference, OpenAIChatCompletionRequest
 from llama_stack.apis.scoring import ScoringResultRow
 from llama_stack.apis.scoring_functions import ScoringFnParams
 from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
@@ -55,7 +55,7 @@ class LlmAsJudgeScoringFn(RegisteredBaseScoringFn):
             generated_answer=generated_answer,
         )
 
-        judge_response = await self.inference_api.openai_chat_completion(
+        params = OpenAIChatCompletionRequest(
             model=fn_def.params.judge_model,
             messages=[
                 {
@@ -64,6 +64,7 @@ class LlmAsJudgeScoringFn(RegisteredBaseScoringFn):
                 }
             ],
         )
+        judge_response = await self.inference_api.openai_chat_completion(params)
         content = judge_response.choices[0].message.content
         rating_regexes = fn_def.params.judge_score_regexes