Using an Agent in gemini enterprise whenever submit a StreamAssistRequest

### Determine this is the right repository

- [x] I determined this is the correct repository in which to report this feature request.

### Summary of the feature request

I would like to easily be able to use an agent for my stream assist request in discovery agent, but seems to be impossible using the actual client libraries

`
from google.api_core.client_options import ClientOptions
from google.cloud import discoveryengine_v1 as discoveryengine

project_id = "myproject"
location = "eu"
engine_id = "myEngine"
search_query = "Ciao come ti chiami?"

def stream_assist_sample(
project_id: str,
location: str,
engine_id: str,
query: str,
):
client_options = (
ClientOptions(api_endpoint=f"{location}-discoveryengine.googleapis.com")
if location != "global"
else None
)

client = discoveryengine.AssistantServiceClient(client_options=client_options)

request = discoveryengine.StreamAssistRequest(
name=client.assistant_path(
project=project_id,
location=location,
collection="default_collection",
engine=engine_id,
assistant="default_assistant",
),
query=discoveryengine.Query(text=query),
)

stream = client.stream_assist(request=request)

for response in stream:
print(response)

stream_assist_sample(project_id, location, engine_id, search_query)`

### Additional context

I'm not able to make a StreamAssistRequest specifying an agent neither from the REST API, it seems to let me choose an AgentsSpec, but it completely ignores it

Start with AssistantServiceClient.stream_assist and the StreamAssistRequest example in this issue, then compare the client-library request with the REST API's AgentsSpec behavior. Determine how an agent should be specified and whether the request must preserve that selection; done means a stream assist request can select an agent through the client library and the behavior is covered by an appropriate test.

Escrita pelo modelo de indexação a partir do texto da issue.