Specifying bucket names in the backup credentials file

There's currently a default bucket name for backups that is hardcoded in the backup code, with an option to override it in the bucket parameters. For users who cannot use the global default, this forces them to specify it for every single backup. This can be especially complicated if they are using multiple backup accounts and need to use different buckets for different accounts. To help support these kind of users, could we add a field in the backup credentials files to specify the bucket name for each account, and use that account name instead of the global default when no bucket name is specified in the backup parameters?

Start by locating the backup credentials-file parser and the code that resolves bucket parameters against the global default. Trace how multiple backup accounts are represented and check existing backup tests. Done means an account-level bucket is read from the credentials file and is used when backup parameters omit a bucket, while an explicit parameter still takes precedence.

Written by the indexing model from the issue text.