Flows: Automatically pause flow schedules if dependent tasks are failing

[zaychenko-sergei]

When a task associated with a flow is permanently failing (assuming retry logic is implemented by this moment), we should not waste resources to schedule next attempts.

Instead:

• flow configuration must be automatically paused
• addt a public event for permanently failing flow

In future, we will be sending e-mails on the platform indicating something is wrong with the flow, and it was paused.