I ran your sketch… then belatedly looked at it when nothing much seemed to happen 
I still think it is bad form to program with delay() of any significant duration as that simply shuts down the “thinking” and doesn’t really equate to real work in an example, sure one core thinks while another sleeps… but what about the alluded one core works hard, while another works harder… without stepping on each others toes?
BTW I do like the way you showed the pinning options in the example… I will have to invest more time to fully get my sleepy head around it.
But what I really like it for is the foundation of a real world example of controlling a stepper from Blynk, running on one core, with the stepper code on the 2nd… at least that is what I am challenging myself to make from your tutorial. It should be a nice upgrade from my old-skool dual MCU