fixed routing table update intervals miss changes

I'm not too thrilled with this change...it feels messier...but it's a
start at fetching Routing Tables on demand and caching them for some
period of time. This solves the problem of a new database being
created and it not being accessible immediately.

There are still some issues for newly created databases and they need
to be ironed out, but existing databases should still behave the same
as before.

horrible things! make proxy work with the BetterSolution(tm)

Aka the current SSR config that's being tested in GKE

does your proxy make TOO MUCH NOISE?!

Clean up logging a bit, having 3 loggers (info, debug, warn) that
allow the user to control the logging detail. Now most of the noise is
hidden behind a -debug flag or a BOLT_PROXY_DEBUG environment
variable.

Uncovered some areas that need some cleanup. Joy.

occasional channel issues on disconnect events

Really really really need to clean this up soon...for now just make
new channels when we switch backend server connections.