fix: ensure assistant message is followed by tool call message as expected by openai (#3224)

# What does this PR do?

As described in #3134 a langchain example works against openai's
responses impl, but not against llama stack's. This turned out to be due
to the order of the inputs. The langchain example has the two function
call outputs first, followed by each call result in turn. This seems to
be valid as it is accepted by openai's impl. However in llama stack,
these inputs are converted to chat completion inputs and the resulting
order for that api is not accpeted by openai.

This PR fixes the issue by ensuring that the converted chat completions
inputs are in the expected order.

Closes #3134 

## Test Plan
Added unit and integration tests. Verified this fixes original issue as
reported.

---------

Signed-off-by: Gordon Sim <gsim@redhat.com>

llama_stack/providers/inline/agents/meta_reference/responses/utils.py

@@ -101,14 +101,22 @@ async def convert_response_input_to_chat_messages(
     """
     messages: list[OpenAIMessageParam] = []
     if isinstance(input, list):
+        # extract all OpenAIResponseInputFunctionToolCallOutput items
+        # so their corresponding OpenAIToolMessageParam instances can
+        # be added immediately following the corresponding
+        # OpenAIAssistantMessageParam
+        tool_call_results = {}
         for input_item in input:
             if isinstance(input_item, OpenAIResponseInputFunctionToolCallOutput):
-                messages.append(
-                    OpenAIToolMessageParam(
-                        content=input_item.output,
-                        tool_call_id=input_item.call_id,
-                    )
+                tool_call_results[input_item.call_id] = OpenAIToolMessageParam(
+                    content=input_item.output,
+                    tool_call_id=input_item.call_id,
                 )
+
+        for input_item in input:
+            if isinstance(input_item, OpenAIResponseInputFunctionToolCallOutput):
+                # skip as these have been extracted and inserted in order
+                pass
             elif isinstance(input_item, OpenAIResponseOutputMessageFunctionToolCall):
                 tool_call = OpenAIChatCompletionToolCall(
                     index=0,
@@ -119,6 +127,9 @@ async def convert_response_input_to_chat_messages(
                     ),
                 )
                 messages.append(OpenAIAssistantMessageParam(tool_calls=[tool_call]))
+                if input_item.call_id in tool_call_results:
+                    messages.append(tool_call_results[input_item.call_id])
+                    del tool_call_results[input_item.call_id]
             elif isinstance(input_item, OpenAIResponseOutputMessageMCPCall):
                 tool_call = OpenAIChatCompletionToolCall(
                     index=0,
@@ -146,6 +157,10 @@ async def convert_response_input_to_chat_messages(
                         f"Llama Stack OpenAI Responses does not yet support message role '{input_item.role}' in this context"
                     )
                 messages.append(message_type(content=content))
+        if len(tool_call_results):
+            raise ValueError(
+                f"Received function_call_output(s) with call_id(s) {tool_call_results.keys()}, but no corresponding function_call"
+            )
     else:
         messages.append(OpenAIUserMessageParam(content=input))
     return messages