[BACKPORT][1.0.1] Allow to preserve OS images when run 'epicli apply' for existing cluster

[seriva]

Backport issues #2288
Related pull requests: #2329

[przemyslavic]

Tested together with #2333.

Tested:
✅ apply v0.9 -> upgrade to v1.0 -> apply v1.0
✅ apply v1.0-> apply v1.0

I checked different values for default_os_image parameter:

• ubuntu-18.04-x86_64

• redhat-7-x86_64

• centos-7-x86_64

• centos-7-arm64 (only terraform part as arm support will be introduced in v1.1.0)

• default

also with changed default specification:

---
kind: infrastructure/virtual-machine
title: "Virtual Machine Infra"
provider: aws
name: default
specification:
  os_full_name: RHEL-7.9_HVM-20210208-x86_64-0-Hourly2-GP2

both for Azure and AWS.