Log model_output_score from Vertex AI response

Per https://cloud.google.com/vertex-ai/docs/generative-ai/model-reference/code-completion, Vertex AI returns a score value for code completion:

A float value that's less than zero. The higher the value for score, the greater confidence the model has in its response.

Logging this field will enable us to track the confidence of the prediction. We might, for example, filter low confidence predictions.