Add exponential back-off retry to reconcile the unhealthy virtual machines in Vsphere

**What happened**:
In our EKS Anywhere cluster, a virtual machine in the vsphere failed with a kernel panic. As part of reconcile process EKS Anywhere tried to power off the machine and destroy the machine. For some reason, EKS Anywhere is not able to power off the troubled machine and constantly retried to Power off the machine.

We expect the reconciliation process to be retried with an exponential back-off option.

**What you expected to happen**:
We expect the reconciliation process to be retried with an exponential back-off option.

**How to reproduce it (as minimally and precisely as possible)**:
Our team updated the steps to reproduce in a separate forum with EKS Anywhere Team

- EKS Distro Release: v1.23.14-eks-1-23-10

Please let me know if you need any further information.

No files, tests, or entry points are named, and the reproduction steps are in a separate forum. Start by locating the vSphere virtual-machine reconciliation path and its handling of failed power-off operations; done means retries use exponential back-off rather than retrying constantly.

Written by the indexing model from the issue text.