Support subsystem code memory experiments

[perlinm]

I think the main challenge for this issue is figuring out how how to (best) combine parity check measurements into stabilizer detectors.

[perlinm]

The choice of measurement schedule must also be made more carefully for subsystem codes, since simply measuring all gauge operators in a random order may not provide enough information to construct a complete set of detectors.

[perlinm]

See comment for supporting CSS codes