Revert backup.sh change, fix README to match actual log output

Author: tvancott42

README.md

@@ -193,7 +193,7 @@ Files are stored canonically by camera, with date-based symlinks for browsing by
 - **Container shows unhealthy**: This is normal until the first successful backup completes. With `RUN_ON_START=true` (the default), this resolves within a few minutes of starting.
 - **Disk space**: The backup script logs two levels of disk space warnings that you can use to set up alerts (e.g., Docker log monitoring, webhook, etc.):
   - `[backup] ... WARN: Archive volume has less than 100 GB free` - time to plan cleanup
-  - `[backup] ... CRITICAL: Archive volume has less than 10 GB free` - backups may start failing
+  - `[backup] ... WARN: CRITICAL: Archive volume has less than 10 GB free` - backups may start failing
 
   There is no automatic pruning yet (coming soon). You are responsible for managing retention (e.g., deleting old date folders, NAS-level quotas, or a cron job). Staging is cleaned after each run.
 - **Permissions**: The installer needs Docker access and write permissions to the archive path - on most NAS systems you're already root. The container runs as root internally for cron, SSH key handling, and volume writes. It does not expose any ports or accept inbound connections.

scripts/backup.sh

@@ -52,7 +52,7 @@ mkdir -p "${STAGING_DIR}/ubv" "$REMUX_DIR"
 # ── Disk space check ────────────────────────────────────────────────────────
 archive_avail_kb=$(df -k "$ARCHIVE_DIR" 2>/dev/null | tail -1 | awk '{print $4}')
 if [ "${archive_avail_kb:-0}" -lt 10485760 ]; then
-    log "CRITICAL: Archive volume has less than 10 GB free — backups may fail"
+    warn "CRITICAL: Archive volume has less than 10 GB free — backups may fail"
 elif [ "${archive_avail_kb:-0}" -lt 104857600 ]; then
     warn "Archive volume has less than 100 GB free — consider pruning old recordings"
 fi