Create basic monitoring setup for the lambdas

I actually have no idea on what we need and can implement without much effort. My ideas so far include:
- enabling x-ray
- maybe some common metrics like error rate?
- slack integration is probably our best bet for visibility -- no other place will be checked
- would be great to have a thought-out workflow for testing production issues
- monitoring for database?