How to delete a state machine?

Complementary to the feature introduced in SB-175 Schedule serverless tasks for specific date, this script was introduced.

If an environment is destroyed, the state machine responsible for task scheduling won't be destroyed until any of its executions have state RUNNING. In the case of those WaitForDueDate executions, that means waiting for weeks until it happens, which can effectively block redeploy of the specific environment.

To stop all RUNNING executions in the AWS Console, a user would have to manually stop each of the execution one after another: there's no easy way to stop all of them in batch, and if there're a few thousands of the RUNNING executions (which proved to be highly possible), it would be a highly time-consuming task.

The script automates the above.