chore(apis): unpublish deprecated /v1/inference apis (#3297)

# What does this PR do?

unpublish (make unavailable to users) the following apis -
 - `/v1/inference/completion`, replaced by `/v1/openai/v1/completions`
- `/v1/inference/chat-completion`, replaced by
`/v1/openai/v1/chat/completions`
 - `/v1/inference/embeddings`, replaced by `/v1/openai/v1/embeddings`
 - `/v1/inference/batch-completion`, replaced by `/v1/openai/v1/batches`
- `/v1/inference/batch-chat-completion`, replaced by
`/v1/openai/v1/batches`

note: the implementations are still available for internal use, e.g.
agents uses chat-completion.

llama_stack/apis/inference/inference.py

@@ -1008,7 +1008,6 @@ class InferenceProvider(Protocol):
 
     model_store: ModelStore | None = None
 
-    @webmethod(route="/inference/completion", method="POST", level=LLAMA_STACK_API_V1)
     async def completion(
         self,
         model_id: str,

llama_stack/providers/utils/inference/prompt_adapter.py

@@ -192,6 +192,14 @@ async def localize_image_content(uri: str) -> tuple[bytes, str] | None:
                 format = "png"
 
         return content, format
+    elif uri.startswith("data"):
+        # data:image/{format};base64,{data}
+        match = re.match(r"data:image/(\w+);base64,(.+)", uri)
+        if not match:
+            raise ValueError(f"Invalid data URL format, {uri[:40]}...")
+        fmt, image_data = match.groups()
+        content = base64.b64decode(image_data)
+        return content, fmt
     else:
         return None