fix: actually propagate inference metrics

currently metrics are only propagated in `chat_completion` and `completion`

since most providers use the openai_.. routes as the default in llama-stack-client inference chat-completion, metrics are currently not working as expected.

in order to get them working the following had to be done:

1. get the completion as usual
2. use new openai_ versions of the metric gathering functions which use .usage from the OpenAI.. response types to gather the metrics which are already populated.
3. define a `stream_generator` which counts the tokens and computes the metrics
4. use a NEW span and log_metrics because the span of the request ends before this processing is complete, leading to no logging unless a custom span is used
5. add metrics to response

Signed-off-by: Charlie Doern <cdoern@redhat.com>

llama_stack/providers/utils/telemetry/tracing.py

@@ -81,7 +81,7 @@ BACKGROUND_LOGGER = None
 
 
 class BackgroundLogger:
-    def __init__(self, api: Telemetry, capacity: int = 1000):
+    def __init__(self, api: Telemetry, capacity: int = 100000):
         self.api = api
         self.log_queue = queue.Queue(maxsize=capacity)
         self.worker_thread = threading.Thread(target=self._process_logs, daemon=True)