feature: apply lambda=1 in the timestep that there is no value output

[YuriCat]

For training involving steps where the outputted value does not exist, it is necessary to set lambda to 1 locally.

I am not sure if VTrace is correct in this.

[YuriCat]

This is ugly and complicated, I think...

[YuriCat]

VTrace looks working well in experiments with Geister.