1. Proposed title of this feature request
Fixed number of retries if Virtual Machine PowerOn Fails

2. What is the nature and description of the request?
When scaling up a new node within an IPI cluster on VMware infrastructure, it's possible for the PowerOn of the new virtual machine to fail for recoverable reasons.

Right now, the PowerOn call is only attempted once, and if it fails, it stalls the scale up process.

In a number of circumstances, a manual poweron starts the VM without issues and the scale up continues.

3. Why does the customer need this? (List the business requirements here)
Makes scale up / powering on a new VM more reliable and eases burden investigating why a scale up failed.

4. List any affected packages or components.
machine-api