Refactor tag overrides

[Alex-Welsh]

This refactor includes two changes

The first sets kolla_base_distro according to ansible facts, and is resolved by kolla-ansible on a host by host basis.

The second reformats the existing structure for override tags, again allowing kolla-ansible to resolve them on a host by host basis.

The result of these two changes is that clouds with with mixed host distributions can be deployed.

This is relevant when performing a rolling migration between two distributions.

[Alex-Welsh]

I've been using this to test the migration from CS8 to RL9. I'm not sure if it needs to be merged into our main branches given that we also have mark's full tag refactor on the horizon. https://github.com/stackhpc/stackhpc-kayobe-config/tree/wallaby-kolla-build-improvements

[markgoddard]

Missing octavia from https://github.com/stackhpc/stackhpc-kayobe-config/pull/524/files

[Alex-Welsh]

Missing octavia from https://github.com/stackhpc/stackhpc-kayobe-config/pull/524/files

Fixed

[markgoddard]

pep8 doesn't like it:

ValueError: 94b83c151902cd5f: found several files in one commit (b'd30e726c573429633e851a71bb619f48821ae83a') with the same UID: ['refactor-tag-overrides-94b83c151902cd5f.yaml', 'refactor-tag-overrides-94b83c151902cd5f.yaml']

[Alex-Welsh]

Tox has stopped complaining now. Not sure what happened there

[Alex-Welsh]

This PR is cursed

[markgoddard]

Rocky 8 failing on MariaDB unrelated:

TASK [mariadb : Wait for MariaDB service to be ready through VIP] **************
Thursday 25 May 2023  10:44:21 +0000 (0:00:03.279)       0:06:39.757 ********** 
FAILED - RETRYING: Wait for MariaDB service to be ready through VIP (6 retries left).
FAILED - RETRYING: Wait for MariaDB service to be ready through VIP (5 retries left).
FAILED - RETRYING: Wait for MariaDB service to be ready through VIP (4 retries left).
FAILED - RETRYING: Wait for MariaDB service to be ready through VIP (3 retries left).
FAILED - RETRYING: Wait for MariaDB service to be ready through VIP (2 retries left).
FAILED - RETRYING: Wait for MariaDB service to be ready through VIP (1 retries left).
fatal: [controller0]: FAILED! => changed=false 
  attempts: 6
  cmd:
  - docker
  - exec
  - mariadb
  - mysql
  - -h
  - 192.168.33.2
  - -P
  - '3306'
  - -u
  - root
  - -piDULu9XXb9VssTIVEEml41mUr2eldZBroLJUqzX8
  - -e
  - show databases;
  delta: '0:00:00.303536'
  end: '2023-05-25 10:45:30.898725'
  msg: non-zero return code
  rc: 1
  start: '2023-05-25 10:45:30.595189'
  stderr: 'ERROR 2013 (HY000): Lost connection to server at ''handshake: reading initial communication packet'', system error: 11'
  stderr_lines: <omitted>
  stdout: ''
  stdout_lines: <omitted>