Investigate migrating bots to serverless function calls instead of DO k8s

[0xAurelius]

To reduce running costs and deployment complexity, these bots could be migrated to a set of serverless function calls

As long as the serverless platform supports Python function calls on a sub-minute latency, it would likely be much cheaper than the current k8s setup, though that should be confirmed with some cost estimation with the planned function provider

Simple multiply the number of function calls per minute (scaled by the amount of resources required) by the number of updates per month (most bots update every few minutes or once per minute)

The key question is whether modern serverless hosting providers can provide sub-minute responses at lower cost than the current setup for all current bots and any projected new bots. Python typically has slow boot times so this likely means either "warm" function calls, or potentially redeveloping these all in TypeScript for consistency with the rest of the codebase (with the help of ChatGPT?)

Scoping this as an investigation ticket to make it achievable, if this were to be implemented there would need to be subsequent tickets