QickSweep can now calclate the actual rounded sweep points

[yoshi74ls181]

Added QickSweep.get_actual_values, which calculates the actual sweep points after rounding to ASM units. There are also some unrelated fixes:

• Documented the keyword argument ro_ch for QickProgramV2.add_pulse
• Fixed a typo in pyro.make_proxy

[meeg]

Looks good, thanks!

I am a little worried that QickSweep.get_actual_values() is largely redundant with QickProgramV2.get_pulse_param() but they both work, I don't want to get in your way, and your solution might be better. I can think about it later, and for now I am happy to merge this.

[yoshi74ls181]

Thank you very much for the quick merge!

I am a little worried that QickSweep.get_actual_values() is largely redundant with QickProgramV2.get_pulse_param()

The advantage of the new QickSweep.get_actual_values() is that it works for a Delay as well as for a Pulse. A disadvantage might be that its implementation involves recursions and closures. If that is not a big concern, maybe it's a good idea to rewrite QickProgramV2.get_pulse_param() to use the new QickSweep.get_actual_values(). The former will still be useful because the user doesn't need to pass to it the loop order and numbers of iterations.

[meeg]

Yes, agree that it's important to be able to get rounded delays. My idea for that was that you can specify an arbitrary "tag" string as a parameter to a timed instruction (pulse, trigger, delay, wait), and then you can do something like QickProgramV2.get_time_param(tag). Does that sound good?

If get_pulse_param() and get_time_param() are sufficient I would want to make those the preferred user interface, because they work with scalar values as well as sweeps. And I agree, probably both should use get_actual_values() internally, I think it is a better design than what I had.

[yoshi74ls181]

Your idea for get_time_param(tag) sounds good! get_pulse_param() and get_time_param() can be the preferred user interface when using qick_lib directly, but for writing our QCoDeS driver QickSweep.get_actual_values() makes more sense because it works in the same way for pulses and delays.