Configure Sentry for error/crash report aggregation

[motin]

This supports resolving #255

Docs: https://sentry.io/for/python/, https://github.com/getsentry/sentry-python

[motin]

With PR #415, activation of Sentry works fine (tried in a GCP crawl) and uncaught exceptions are caught. This at least allows us to get some notification on uncaught exceptions in a crawl.

In an additional PR, I hope that we can also report the caught exceptions to sentry so that we can get a good overview of what is causing driver-related crashes in the crawls.

Also, integrating with the OpenWPM logger (eg sending log messages to Sentry as well as per https://docs.sentry.io/platforms/python/logging/#usage) allows accessing the relevant log scrollback for each reported exception.

@englehardt Care to set that up? I am not sure where to best add that code.

[motin]

Possibly useful for crawls as well, adding Redis commands as breadcrumbs: https://docs.sentry.io/platforms/python/redis/