Async node update

[stefanopepe]

As a Validator, I need an asynchronous node update process, such that I can script the migration process

Professional validators use scripting and infrastructure as code, so they need to plan in advance any migration that will happen to the network. Ideally, they need a "rendezvous point" (a block height) within their new release of the node should ready to take the lead. This card adds details to #166

Acceptance Criteria (or use this as a ToDo):

• [ ] Document the process of switching validator nodes without downtime (clone of #341)
• [ ] Document the tooling we use for state migration, in case of a planned hard fork

What is Out of Scope:

• Discussing if we will ever have a hard fork in the future. If it will happen, this thing might be useful

[amgando]

let's capture this process being completed by an expert so there's clarity on what to document

[stefanopepe]

I suppose we need to identify such an expert. @bowenwang1996 do you have any ideas?

[bowenwang1996]

I can write documentation for this.

[chadoh]

Is this purely a docs issue?