Improve the doc for HA

The current document for HA does not contain the following details such as:

1. Architecture Overview

2. High Availability Strategy

3. Failover and Redundancy

4. Backup and Recovery

5. Upgrade strategy

   1. Can it be done in a rolling fashion?

      1. Preliminary input from Nurlan: Good question, theoretically yes. Depending on migrations type there might be some downtime.

6. Shutdown strategy

7. Migration

   1. How to migrate from PMM "single instance" to PMM in "HA mode" keeping the current monitoring data - if possible

      1. Current guide is about creating a new fresh HA environment, but one common use case is migration from single-mode PMM so we will need to be prepared on this topic

8. Replacement of a PMM node

   1. A node where PMM is running is down and it will not be available anymore (hardware failure). How to add a new node to the already existing "cluster"

      1. Preliminary input from Nurlan: not tested yet, but it should be possible just by adding one more instance with changed node addresses.

9. Shutdown

   1. Any specific order/sequence for the shutdown (say I want to shutdown all services)

      1. Preliminary input from Nurlan: no, it can be shutdown in any order. But the recommended one is 1. PMM passives 2. PMM active 3. HAProxy and DBs.

10. How to start the external DBs in HA mode

    1. More low priority and outdated now that the helm chart does all the work.

    2. Do we need something for the manual mode?

Add these sections to the document.