ci(mint): set RUSTFS_UNSAFE_BYPASS_DISK_CHECK so the mint container boots (#4814)

ci(mint): bypass local physical-disk-independence guard so mint boots

The mint container maps four RUSTFS_VOLUMES data dirs onto a single
runner device, so the startup physical-disk-independence guard aborts
with a FATAL before mint can run. The scheduled run 29183544431
(2026-07-12) crashed at 'Wait for RustFS ready' with 'local erasure
endpoints must use distinct physical disks'.

Set RUSTFS_UNSAFE_BYPASS_DISK_CHECK=true on the container, the CI use
the guard explicitly sanctions -- mirroring the e2e-s3tests harness
fixed in #4768. Completes the ci-2 acceptance (a mint run that actually
produces log.json and the per-suite summary).

.github/workflows/mint.yml

@@ -135,6 +135,10 @@ jobs:
         run: |
           docker network inspect rustfs-net >/dev/null 2>&1 || docker network create rustfs-net
           docker rm -f rustfs-mint >/dev/null 2>&1 || true
+          # The four disks share one physical device on the runner (a single
+          # loopback filesystem), so the local physical-disk-independence guard
+          # would refuse to start. Bypass it — this is the CI use case the guard
+          # explicitly sanctions via RUSTFS_UNSAFE_BYPASS_DISK_CHECK.
           docker run -d --name rustfs-mint \
             --network rustfs-net \
             -p 9000:9000 \
@@ -142,6 +146,7 @@ jobs:
             -e RUSTFS_ACCESS_KEY="${S3_ACCESS_KEY}" \
             -e RUSTFS_SECRET_KEY="${S3_SECRET_KEY}" \
             -e RUSTFS_VOLUMES="/data/rustfs{0...3}" \
+            -e RUSTFS_UNSAFE_BYPASS_DISK_CHECK=true \
             -v /tmp/rustfs-mint:/data \
             rustfs-ci