fix: move jellyfin config to local /var/lib/jellyfin (NFS SQLite lock stalls); decommission OpenVAS

Jellyfin lockups root-caused to jellyfin.db (WAL mode) on NFS: SQLite
"database is locked" errors escalated 22->33->125/day, with DB commands
blocking 60-102s, causing healthcheck flaps and playback stalls. Config
rsynced from /mnt/docker-data/configs/jellyfin to local /var/lib/jellyfin
on bart; healthy after redeploy. jellyfin:2026060106 v12 image pin
unchanged. Old NFS copy left in place as backup.

OpenVAS fully removed from bart per user (~39GB reclaimed: containers,
volumes incl. PG DB, image). Prometheus probe and Farm Manager link
removed. Compose dir kept for reference; Known Issue #21 marked moot.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

compose/bart/media/docker-compose.yml

@@ -13,7 +13,10 @@ services:
       - PGID=1000
       - LIBVA_DRIVER_NAME=i965   # Haswell HD 4600 uses legacy i965 (iHD is Broadwell+)
     volumes:
-      - /mnt/docker-data/configs/jellyfin:/config
+      # LOCAL disk, not NFS: jellyfin.db (SQLite WAL) on NFS caused 60-100s
+      # "database is locked" stalls -> server lockups during playback
+      # (2026-07-19). Old copy left at /mnt/docker-data/configs/jellyfin.
+      - /var/lib/jellyfin:/config
       - /mnt/nas-media:/media
     ports:
       - "8096:8096"