[Task] Store reports in S3

[andrewazores]

See #9 #51 #59 . Reports are currently generated and held in-memory in a Caffeine cache. Reports are generally small so this is OK for the most part, but over time the number of reports may exert noticeable memory pressure, so some cache tuning needs to be applied. Also, reports for archived recordings are likely to be accessed infrequently, but it is still worth saving the cost of regenerating a report (requires intensive CPU and memory to generate, but only takes a very small amount of disk to store), so they should be tiered between in-memory for rapid access and in S3 alongside the source recording for later reference, or for access after the Cryostat server is restarted.

[andrewazores]

Fixed in #104