[BACKPORT][v1.5.5][REFACTOR] move mount point check function to common lib

[github-actions]

backport #7394

[longhorn-io-github-bot]

Pre Ready-For-Testing Checklist

• Where is the reproduce steps/test steps documented?
  The reproduce steps/test steps are at:

1. Create single node cluster curl -sfL https://get.k3s.io | INSTALL_K3S_VERSION=v1.28.5+k3s1 sh -s - --write-kubeconfig-mode 644
2. Deploy Longhorn master-head
3. Create a single replica volume, attach to node with deployment, record the PVC name
4. Find the PVC mount location mount | grep "/dev/longhorn/${VOLUME_NAME}"
/host/var/lib/kubelet/plugins/kubernetes.io/csi/driver.longhorn.io/${VOLUME_NAME_HASH}/globalmount
5. Remount PVC to read only by below command (remember the path ends with "globalmount" )

mount -o remount,ro /host/var/lib/kubelet/plugins/kubernetes.io/csi/driver.longhorn.io/${VOLUME_NAME_HASH}/globalmount

6. The engine condition should back to FaileSystemReadOnly: false and the IO works in the workload

PRs:
longhorn/go-common-libs#19
longhorn/longhorn-instance-manager#428
longhorn/longhorn-share-manager#152

[roger-ryao]

Verified on v1.5.x-head 20240327

• longhorn v1.5.x-head 3df6f3b
• longhorn-share-manager v1.5.x-head longhorn/longhorn-share-manager@1d4afd5
• longhorn-instance-manager v1.5.x-head longhorn/longhorn-instance-manager@7cd3a54

The test steps

#8125 (comment)

Result Passed

The volume's engine condition would change back to FailedSystemReadOnly: false, and the IO would work in the workload.

> k -n longhorn-system describe engine vol-0-e-0 |grep -H3 'FilesystemReadOnly'
(standard input)-    Message:               
(standard input)-    Reason:                
(standard input)-    Status:                False
(standard input):    Type:                  FilesystemReadOnly
(standard input)-  Current Image:           longhornio/longhorn-engine:v1.5.x-head
(standard input)-  Current Replica Address Map:
(standard input)-    vol-0-r-b1b49dcf:        10.42.0.21:10000