Feat: Add s390x rhel support

[AmadeusPodvratnik]

This is the fix for ticket:
#790

Problem:
Using ceph-cm-ansible is causing issues with RHEL especially on s390x hypervisors like zVM.
Different packages needed for s390x architecture.

Fix:
Use ansible_facts.architecture as an architecture flag to load the correct software.
Disable hwclock and /etc/grub.d/02_force_timeout.

Packages names are adapted to fit into rhel on s390x.

rhel-9-for-s390x-baseos-e4s-rpms
rhel-9-for-s390x-appstream-eus-rpms
codeready-builder-for-rhel-9-s390x-rpms

Test:

The fix was tested on s390x (zVM) using RHEL
cat /etc/os-release to
NAME="Red Hat Enterprise Linux"
VERSION="9.6 (Plow)"
ID="rhel"

ansible-playbook /home/teuthology/ceph-cm-ansible/cephlab.yml -vvv --extra-vars '{"ansible_ssh_user": "ubuntu"}' -i /tmp/testnode_inventory

PLAY RECAP ***********************************************************************************************************************************************************************************************************************
a3e11001.lnxero1.boe : ok=124 changed=36 unreachable=0 failed=0 skipped=247 rescued=0 ignored=3
a3e11008.lnxero1.boe : ok=124 changed=36 unreachable=0 failed=0 skipped=247 rescued=0 ignored=3

[djgalloway]

Looks like you removed then re-added

# Set to false if disk should not cleaned - simple runs for nodes with one disk only.
wipefs_disk: true

In two separate commits. If you just squash your commits, it should clear up that condition.