feat(responses)!: improve responses + conversations implementations (#3810)

This PR updates the Conversation item related types and improves a
couple critical parts of the implemenation:

- it creates a streaming output item for the final assistant message
output by
  the model. until now we only added content parts and included that
  message in the final response.

- rewrites the conversation update code completely to account for items
  other than messages (tool calls, outputs, etc.)

## Test Plan

Used the test script from
https://github.com/llamastack/llama-stack-client-python/pull/281 for
this

```
TEST_API_BASE_URL=http://localhost:8321/v1 \
  pytest tests/integration/test_agent_turn_step_events.py::test_client_side_function_tool -xvs
```

tests/integration/responses/helpers.py

@@ -16,21 +16,11 @@ def new_vector_store(openai_client, name, embedding_model, embedding_dimension):
             openai_client.vector_stores.delete(vector_store_id=vector_store.id)
 
     # Create a new vector store
-    # OpenAI SDK client uses extra_body for non-standard parameters
-    from openai import OpenAI
-
-    if isinstance(openai_client, OpenAI):
-        # OpenAI SDK client - use extra_body
-        vector_store = openai_client.vector_stores.create(
-            name=name,
-            extra_body={"embedding_model": embedding_model, "embedding_dimension": embedding_dimension},
-        )
-    else:
-        # LlamaStack client - direct parameter
-        vector_store = openai_client.vector_stores.create(
-            name=name, embedding_model=embedding_model, embedding_dimension=embedding_dimension
-        )
 
+    vector_store = openai_client.vector_stores.create(
+        name=name,
+        extra_body={"embedding_model": embedding_model, "embedding_dimension": embedding_dimension},
+    )
     return vector_store