[Task] — Storage compose scripts: no `set -e`, no trap, hardcoded sleeps

- severity: Medium; files: same four `e2e-*-compose.sh` storage scripts
- description: No `set -euo pipefail` (the `http` compose script has it; these don't). No `trap` to run `docker compose down` on failure/exit. Uses `sleep 30s` then `sleep 60s` instead of polling for healthy.
- impact: A failed `storage_init_*.sh` or `docker compose up` is silently ignored; services left running after the job.
- suggested_fix: Add `set -euo pipefail`; add `trap 'docker compose -f ... down' EXIT`; replace `sleep` with a poll-until-healthy loop.
- confidence: High

---
_Identified during the 2026-08-02 deep re-scan; full list in [`docs/scan2-2026-08-02/06-medium-tiers.md`](docs/scan2-2026-08-02/06-medium-tiers.md)._

Start by comparing the four storage e2e-*-compose.sh scripts with the HTTP compose script, then inspect the referenced storage_init_*.sh commands and the scan notes in docs/scan2-2026-08-02/06-medium-tiers.md. Run the storage compose scripts to verify failures stop the job, cleanup runs on exit, and service readiness is polled rather than governed by fixed sleeps.

Written by the indexing model from the issue text.