Consolidate the different implementations of state tomography

[viathor]

We have three implementations of state tomography:

• for 1-qubit states,
• for 2-qubit states,
• for n-qubit states for arbitrary n.

The last one isn't well tested (yet). Once it is, we should compare performance and if there isn't a clear advantage to the fixed-size implementations we should remove them.

[mpharrigan]

Should we keep this open? The linked issue fixes the bug in n-qubit tomography but the larger issue is still present

[viathor]

Yes, thank you. My comment on PR #2739 incorrectly said that it fixed #2737. It should have said that it fixed #2738.

[dabacon]

There can never be too many implementations of state tomography :)

But yes we should unify these standard method of state tomography (there are many others).

[MichaelBroughton]

Is this still an open concern ?