feat(proxy_server.py): enable llm api based prompt injection checks

run user calls through an llm api to check for prompt injection attacks. This happens in parallel to th
e actual llm call using `async_moderation_hook`

enterprise/enterprise_hooks/google_text_moderation.py

@@ -96,6 +96,9 @@ class _ENTERPRISE_GoogleTextModeration(CustomLogger):
     async def async_moderation_hook(
         self,
         data: dict,
+        call_type: (
+            Literal["completion"] | Literal["embeddings"] | Literal["image_generation"]
+        ),
     ):
         """
         - Calls Google's Text Moderation API

enterprise/enterprise_hooks/llama_guard.py

@@ -99,6 +99,9 @@ class _ENTERPRISE_LlamaGuard(CustomLogger):
     async def async_moderation_hook(
         self,
         data: dict,
+        call_type: (
+            Literal["completion"] | Literal["embeddings"] | Literal["image_generation"]
+        ),
     ):
         """
         - Calls the Llama Guard Endpoint

enterprise/enterprise_hooks/llm_guard.py

@@ -22,6 +22,7 @@ from litellm.utils import (
 )
 from datetime import datetime
 import aiohttp, asyncio
+from litellm.utils import get_formatted_prompt
 
 litellm.set_verbose = True
 
@@ -94,6 +95,9 @@ class _ENTERPRISE_LLMGuard(CustomLogger):
     async def async_moderation_hook(
         self,
         data: dict,
+        call_type: (
+            Literal["completion"] | Literal["embeddings"] | Literal["image_generation"]
+        ),
     ):
         """
         - Calls the LLM Guard Endpoint