Initial feedback on Wrapper primitives

[jyu00]

What is the expected enhancement?

I just started using Wrapper primitives for the first time, and here's some feedback before I forget:

• It'd be nice if WrapperSampler also return all the input options in result metadata, to match that of WrapperEstimator (not high priority)
• It'd be nice if WrapperEstimator can also return ExecutionSpans (not high priority)
• WrapperEstimator logging says "Submitting 1 pub to executor with 128 shots", which is a bit misleading since it's really 128 shots per randomization.
• My code used TREX and I expected some logging for post-processing but there wasn't any.
• Right now it looks like find_unique_layers() returns both 2Q and measurement layers, and the user needs to strip out measurement layer based on (the lack of) InjectNoise annotation. It'd be less confusing to just have a find_unique_2q_layers() method.

Acceptance criteria

See above.

[SamFerracin]

Thanks, @jyu00 !
We'll go through these, and if you have more inputs, please add them here!

As per private discussion, perhaps let's not take action on the last one ("Right now it looks like find_unique_layers() returns both 2Q and measurement layers, ..") until we have a workflow sketched out for the local-mode, to make sure that we have the full picture in mind