chore: mypy violations cleanup for inline::{telemetry,tool_runtime,vector_io} (#1711)

# What does this PR do?

Clean up mypy violations for inline::{telemetry,tool_runtime,vector_io}.
This also makes API accept a tool call result without any content (like
RAG tool already may produce).

Signed-off-by: Ihar Hrachyshka <ihar.hrachyshka@gmail.com>

llama_stack/providers/inline/tool_runtime/code_interpreter/code_env_prefix.py

@@ -69,7 +69,7 @@ def popen_not_allowed(*args, **kwargs):
     )
 
 
-_subprocess.Popen = popen_not_allowed
+_subprocess.Popen = popen_not_allowed  # type: ignore
 
 
 import atexit as _atexit
@@ -104,7 +104,7 @@ def _open_connections():
     return _NETWORK_CONNECTIONS
 
 
-_builtins._open_connections = _open_connections
+_builtins._open_connections = _open_connections  # type: ignore
 
 
 @_atexit.register

llama_stack/providers/inline/tool_runtime/code_interpreter/code_execution.py

@@ -161,9 +161,9 @@ _set_seeds()\
 def process_matplotlib_response(response, matplotlib_dump_dir: str):
     image_data = response["image_data"]
     # Convert the base64 string to a bytes object
-    images = [base64.b64decode(d["image_base64"]) for d in image_data]
+    images_raw = [base64.b64decode(d["image_base64"]) for d in image_data]
     # Create a list of PIL images from the bytes objects
-    images = [Image.open(BytesIO(img)) for img in images]
+    images = [Image.open(BytesIO(img)) for img in images_raw]
     # Create a list of image paths
     image_paths = []
     for i, img in enumerate(images):

llama_stack/providers/inline/tool_runtime/rag/__init__.py

@@ -11,7 +11,7 @@ from llama_stack.providers.datatypes import Api
 from .config import RagToolRuntimeConfig
 
 
-async def get_provider_impl(config: RagToolRuntimeConfig, deps: Dict[str, Any]):
+async def get_provider_impl(config: RagToolRuntimeConfig, deps: Dict[Api, Any]):
     from .memory import MemoryToolRuntimeImpl
 
     impl = MemoryToolRuntimeImpl(config, deps[Api.vector_io], deps[Api.inference])

llama_stack/providers/inline/tool_runtime/rag/memory.py

@@ -15,6 +15,7 @@ from pydantic import TypeAdapter
 from llama_stack.apis.common.content_types import (
     URL,
     InterleavedContent,
+    InterleavedContentItem,
     TextContentItem,
 )
 from llama_stack.apis.inference import Inference
@@ -23,6 +24,7 @@ from llama_stack.apis.tools import (
     RAGQueryConfig,
     RAGQueryResult,
     RAGToolRuntime,
+    Tool,
     ToolDef,
     ToolInvocationResult,
     ToolParameter,
@@ -62,6 +64,12 @@ class MemoryToolRuntimeImpl(ToolsProtocolPrivate, ToolRuntime, RAGToolRuntime):
     async def shutdown(self):
         pass
 
+    async def register_tool(self, tool: Tool) -> None:
+        pass
+
+    async def unregister_tool(self, tool_id: str) -> None:
+        return
+
     async def insert(
         self,
         documents: List[RAGDocument],
@@ -121,11 +129,11 @@ class MemoryToolRuntimeImpl(ToolsProtocolPrivate, ToolRuntime, RAGToolRuntime):
             return RAGQueryResult(content=None)
 
         # sort by score
-        chunks, scores = zip(*sorted(zip(chunks, scores, strict=False), key=lambda x: x[1], reverse=True), strict=False)
+        chunks, scores = zip(*sorted(zip(chunks, scores, strict=False), key=lambda x: x[1], reverse=True), strict=False)  # type: ignore
         chunks = chunks[: query_config.max_chunks]
 
         tokens = 0
-        picked = [
+        picked: list[InterleavedContentItem] = [
             TextContentItem(
                 text=f"knowledge_search tool found {len(chunks)} chunks:\nBEGIN of knowledge_search tool results.\n"
             )