SentryMonitor Improvements

[calebsyring]

SentryMonitor.__init__ requires an env argument while SentryMonitor.apply_config assumes env from the SENTRY_ENVIRONMENT in the app config. That behavior is inconsistent. If there's no reason for it, let's just make __init__ get env from the app config as well.

It's also annoying to have to pass the org argument in everywhere. Can we make that default to an app config variable (if not force it to come from one, like SENTRY_ENVIRONMENT)?

[calebsyring]

I also had to make some adjustments to our ping task in a client project based on error codes received from Sentry. It's possible we'd want to include those in this lib. I retried 503 errors (which Sentry seems to respond with intermittently). I also ignored 429 rate limiting errors.

Edit: addressed by #193

[calebsyring]

Also looks like Sentry already has some cron functionality in the sentry_sdk package: https://docs.sentry.io/platforms/python/crons. It is possible ours is duplicative? Maybe we only need the config tools for creating monitors.

[calebsyring]

Additionally, it appears monitor slugs must be unique across an organization. We'll want to prepend the project name or otherwise find a way to differentiate monitors between projects.