Description of problem:

With CSI migration enabled && in-tree PV (or inline volume) with a partition specified, the CSI driver tries to mount the whole disk.

Version-Release number of selected component (if applicable):
4.7

How reproducible:

Steps to Reproduce:
1. Enable CSI migration on AWS
2. Manually create a volume and a partition on it.
3. Run this pod:

{
    "apiVersion": "v1",
    "kind": "Pod",
    "metadata": {
        "name": "inline"
    },
    "spec": {
        "nodeSelector": {
                "topology.kubernetes.io/zone": "us-east-2a"
        },
        "containers": [
            {
                "name": "inline",
                "image":
"quay.io/openshifttest/storage@sha256:a05b96d373be86f46e76817487027a7f5b8b5f87c0ac18a246b018df11529b40",
                "securityContext": {
                    "privileged": true
                },
                "imagePullPolicy": "IfNotPresent",
                "volumeMounts": [
                    {
                        "mountPath": "/mnt/storage",
                        "name": "inline"
                    }
                ]
            }
        ],
        "volumes": [
            {
                "name": "inline",
                "awsElasticBlockStore": {
                    "fsType": "xfs",
                    "volumeID": "vol-0140cff45bc3ecdbc",
                    "partition": 1,
                    "readOnly": false
                }
            }
        ]
    }
}


Fix both with "traditional" (volumes are /dev/xvdXY) and Nitro systems (/dev/nvmeXYZ)!

Upstream issue: https://github.com/kubernetes-sigs/aws-ebs-csi-driver/issues/812