Per our discussion, we should also consider testing cases where timeseries range and simulation time range overlap partly. So in total that'd be:

• Not a single timeseries date in simulation time range
• Timeseries starts with a considerable delay after simulation start
• Timeseries ends with a considerable delay before simulation start

There should be tests for all of this.

This problem occurres because of ParticipantAgentFundamantals (lines: 142 - 144). One possible fix is shown below.

current:

senderToMaybeTick._2.foreach { tick =>
    scheduler ! Completion(self.toTyped, Some(tick))
}

fixed:

scheduler ! Completion(self.toTyped, senderToMaybeTick._2)

The bug is fixed with #1016. But we could add some tests.