* low-balance-auto-top-up:
  Some people have exactly 5 who don't need to be told
  On low balance, top-up or notify
  ExpiringLock helper

On start up, check for users with low balance and NOTIFY about them.  LISTEN for
such notifications and process by either sending a low-balance warning message
or else attempting an auto-top-up as configured.

Using NOTIFY/LISTEN because then we can NOTIFY after any INSERT that leaves the
balance too low (using a trigger).  Doing the sync on start-up in case we missed
a NOTIFY during any downtime.  Using the Redis lock to prevent spamming a
low-balance user in case of many restarts or if they have many small
transactions happen in one day.

For things we want to do only so often, set up a helper to push expiring keys to
Redis and check for them.

* cancel-timeout:
  Timeout is not a fatal error
  When user cancels the command, respond with canceled

If the user does not proceed with a command after N time, we don't hold on to it
forever and time out.  We cannot return anything to the user because they
haven't sent us anything, so just ignore it.

If the cancel has not been fully handled in the body of the command, at least
respond with canceled status and do not go to sentry.

* command-object:
  ErrorToSend => FinalStanza
  Since Command#finish causes an error, a then off the end won't work
  Refactor commands to have Command and Command::Execution objects

* statsd-logging:
  Use the logger for all gems

I ran this setup from a semi-clean environment, and there were some
steps I needed to do that weren't captured anywhere.

Show things from all of the start of range (today)

Brings the common elements of all commands together, and threads the most useful
state (such as ability to reply) through automatically using the new EMPromise
fiber trampoline.

Otherwise we won't know when someone has paid...

Using the Repository pattern to encapsulate the fetch and create operations on
the persistence layer for a domain object.  These were not really factories in
the classic sense, but rather "fetch from persisitence layer" methods, and so
they now have a home.

* logging:
  Slightly better logging

Import structured logging gem ougai and replace current use of puts/warn with
logging calls.  Gives us level, timestamp, and formatting for exceptions and/or
hash of arbitrary context data.  Can do JSON logging if we ever need that, but
set to human-readable (with colour if isatty) for now.

* btc_spop:
  Get new Bitcoin address from Redis set

This set is populated by a cron job to have only known-good available addresses
in it.

So it won't be taken as next by naive code.

Instead, adhoc bot will show the options for actions to take

Not on our inbound-facing JID.  This is why we shouldn't be mucking in the SGX's
Redis at all...

Turns out every user will start with no attempts, and the code wasn't
handling that case. Oops.