fix: use memory cache backend for parallel trivy scans

Parallel trivy image workers sharing the fs cache directory hit
"cache may be in use by another process: timeout" errors. Switch to
--cache-backend memory to avoid bolt-db lock contention, matching the
approach already used in generate-sbom.sh.

.github/workflows/cve-edge-post.yml

@@ -103,7 +103,7 @@ jobs:
             img="$1"; reports="$2"
             safe=$(printf "%s" "$img" | tr "/:@" "___")
             echo ">> 스캔: $img"
-            trivy image --quiet --skip-db-update \
+            trivy image --quiet --cache-backend memory --skip-db-update \
               --format json --timeout 15m --output "$reports/$safe.json" "$img" \
               || echo "::warning::스캔 실패: $img"
           ' _ {} "$REPORTS_DIR" < "$OUT_DIR/images_scan.txt"