restic-backup: unlock stale locks before backing up

A backup killed mid-run leaves a lock behind; the next night's backup
still works (non-exclusive) but forget fails and the run reports
rc=1 for ever. restic-maintenance.sh already did this.

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

restic-backup.sh

@@ -149,6 +149,10 @@ fi  # stage dumps
 
 if [ "$STAGE" = all ] || [ "$STAGE" = backup ]; then
 
+# clear locks left by a killed run (safe: only stale ones are removed);
+# otherwise a single interrupted backup blocks forget/prune for ever
+$RESTIC -r "$REPO" unlock >/dev/null 2>&1 || true
+
 log "restic backup -> $REPO"
 $RESTIC -r "$REPO" backup \
     --pack-size 64 --read-concurrency 8 -o s3.connections=8 \