build-ubi keeps failing on basic e2e tests

[tnqn]

Describe the bug

After #5764 was merged, "build-ubi" has never succeeded on main branch: https://github.com/antrea-io/antrea/actions/workflows/build.yml?query=branch%3Amain.

For example: https://github.com/antrea-io/antrea/actions/runs/8226988559/job/22494215561

2024/03/11 02:59:15 Creating K8s ClientSet
2024/03/11 02:59:15 Collecting information about K8s cluster
2024/03/11 02:59:15 Pod IPv4 network: '10.244.0.0/16'
2024/03/11 02:59:15 Service IPv4 network: '10.96.0.0/16'
2024/03/11 02:59:15 Num nodes: 3
2024/03/11 02:59:15 Applying Antrea YAML
2024/03/11 02:59:49 Waiting for all Antrea DaemonSet Pods
2024/03/11 02:59:50 Checking CoreDNS deployment
2024/03/11 03:01:38 Error when deploying Antrea: some CoreDNS replicas are still unavailable after 1m30s
FAIL	antrea.io/antrea/test/e2e	142.824s
FAIL
Cleaning testbed
Deleting cluster "kind" ...
Deleted nodes: ["kind-control-plane" "kind-worker2" "kind-worker"]
Deleting VLAN subnets
Deleting external server
Error: Process completed with exit code 1.

[xliuxu]

The issue is caused by insufficient disk space as we are using the same runner for building and testing. It can be resolved by #6091.