[ENH] - Add one sided asymmetric option to asine cycle

[TomDonoghue]

This update adds support for creating an asymmetric cycle with only one asymmetric extrema, to sim_asine_cycle.

For example, this code:

import matplotlib.pyplot as plt
from neurodsp.sim.cycles import sim_asine_cycle

fs, freq = 1000, 10
cycle = sim_asine_cycle(1/freq, fs, 0.25, 'peak')

plt.plot(cycle)

Would now work, to have a 'peak-only asymmetric cycle', and gives this:

As a limitation of this - it seems unideal that there is an abrupt shift from the asymmetric to symmetric regime at the zero crossing point (ideally this might be smoother), but broadly I think this should be okay, and this update can still be useful for some things.

Edit: I've been playing around with those code for awhile, including poking at different ways to do the offsets, and if there's a way to refactor what looks to be duplicated code. At this point, I think it's probably easiest to stick with it as is, but one way or another I'm moving this to review stage to get some more eyes on it!

[ryanhammonds]

Everything looks good here in terms of functionality. I tested this out in a not book but am still wrapping my head around how the various phase arrays work.

[TomDonoghue]

Thanks for checking @ryanhammonds! I don't think anything necessarily needs any additional figuring out here, so I'm going to call it, and merge this one in!