feat: add batch inference API to llama stack inference (#1945)

# What does this PR do?

This PR adds two methods to the Inference API:
- `batch_completion`
- `batch_chat_completion`

The motivation is for evaluations targeting a local inference engine
(like meta-reference or vllm) where batch APIs provide for a substantial
amount of acceleration.

Why did I not add this to `Api.batch_inference` though? That just
resulted in a _lot_ more book-keeping given the structure of Llama
Stack. Had I done that, I would have needed to create a notion of a
"batch model" resource, setup routing based on that, etc. This does not
sound ideal.

So what's the future of the batch inference API? I am not sure. Maybe we
can keep it for true _asynchronous_ execution. So you can submit
requests, and it can return a Job instance, etc.

## Test Plan

Run meta-reference-gpu using:
```bash
export INFERENCE_MODEL=meta-llama/Llama-4-Scout-17B-16E-Instruct
export INFERENCE_CHECKPOINT_DIR=../checkpoints/Llama-4-Scout-17B-16E-Instruct-20250331210000
export MODEL_PARALLEL_SIZE=4
export MAX_BATCH_SIZE=32
export MAX_SEQ_LEN=6144

LLAMA_MODELS_DEBUG=1 llama stack run meta-reference-gpu
```

Then run the batch inference test case.

llama_stack/providers/inline/inference/meta_reference/parallel_utils.py

@@ -19,7 +19,7 @@ import tempfile
 import time
 import uuid
 from enum import Enum
-from typing import Callable, Generator, Literal, Optional, Union
+from typing import Callable, Generator, List, Literal, Optional, Tuple, Union
 
 import torch
 import zmq
@@ -69,12 +69,12 @@ class CancelSentinel(BaseModel):
 
 class TaskRequest(BaseModel):
     type: Literal[ProcessingMessageName.task_request] = ProcessingMessageName.task_request
-    task: Union[CompletionRequestWithRawContent, ChatCompletionRequestWithRawContent]
+    task: Tuple[str, List[CompletionRequestWithRawContent] | List[ChatCompletionRequestWithRawContent]]
 
 
 class TaskResponse(BaseModel):
     type: Literal[ProcessingMessageName.task_response] = ProcessingMessageName.task_response
-    result: GenerationResult
+    result: List[GenerationResult]
 
 
 class ExceptionResponse(BaseModel):
@@ -331,7 +331,7 @@ class ModelParallelProcessGroup:
 
     def run_inference(
         self,
-        req: Union[CompletionRequestWithRawContent, ChatCompletionRequestWithRawContent],
+        req: Tuple[str, List[CompletionRequestWithRawContent] | List[ChatCompletionRequestWithRawContent]],
     ) -> Generator:
         assert not self.running, "inference already running"