Export concourse audit logs to s3

[cweibel]

Changes proposed in this pull request:

• Drain the audit logs for Concourse to an s3 bucket. Stored in a YYYY/MM/DD/HH/MM/SS folder structure, similar to the cf events scraper that logs-opensearch performs. Files are actually gzipped unlike logs-opensearch.
• Once opensearch for platform logs is available, an ingestor-s3 job can be configured to import them.
• The s3 bucket (and iam creds to be later used by the ingestor), were created with Concourse audit logs bucket & iam user/role terraform-provision#2292.
• The names of the buckets (dev/stage/prod) were copied to Concourse Credhub so they could be used by pipeline.yml in this repo.
• The job runs hourly on a timer. If this interval changes, also need to change the duration logic in export-concourse-audit-logs-to-s3.sh
• Part of https://github.com/cloud-gov/private/issues/2835

security considerations

Secrets are in Credhub. Logs are sent to private/encrypted bucket.

[ChrisMcGowan]