Add job to maintain .repo-metadata information

The information within the `.repo-metdata.json` file for each client library module is initially populated during creation of the new client library. After creation, there is not an automated way for the information to get updated.

The fields should come from the Cloud Drop yaml file and ideally there would be a process such that if the Cloud Drop yaml file is updated, the `.repo-metadata.json` file is also updated.

Many of the fields are used in our external documentation. For instance on the [library overview page](https://cloud.google.com/java/docs/reference/google-cloud-speech/latest/overview), the product documentation URL is explicitly linked, and the `Enable the API` link is created from the `api_shortname`.