[docs] - Example for how to use dagstermill output notebook with distributed storage

[yuhan]

Write an example in documentation to explain how to use dagstermill's output notebook functionality when the job is running in distributed set up (e.g. when the user wants to store the output notebook to a s3 bucket and also potentially have a downstream solid/op accessing to that notebook for use cases like https://github.com/dagster-io/dagster/issues/2319)

Potentially we could improve the output notebook API if we don't feel good about the example code. For example, the changes in https://github.com/dagster-io/dagster/pull/5163 could be awkward to make for remote use cases.