You can restart your cluster using Powershell
https://docs.microsoft.com/en-us/powershell/module/servicefabric/restart-servicefabricnode?view=azureservicefabricps
Or you can restart the cluster from the SF Cluster manager dashboard.
The D drive in Azure is a temp disk. The size will differ depending on the size of the machines. You can see the list of VM sizes here and the associated temp drive size
https://docs.microsoft.com/en-us/azure/virtual-machines/windows/sizes

If you redeploy the node it would allocate a new temp drive and remove all the data from the old one. Which would give you the space back. But the longer term solution would be to allocate a disk drive or store the data on a file share of some sort where
you have control over the drive size. You could also store on the C drive if you prefer you can make that drive larger. But really you should only store data on data disks or file shares to ensure persistence of the data.
Alternatively, you could RDP to the nodes and manually remove the data from the temp drive. This might be the fast solution.