Feature/classical conditionals

[PabloAndresCQ]

Description

Adding support so that simulate can:

• run circuits with classical logic ([StructuredState] Support classical operations and classically controlled gates #151)
• run circuits with a single qubit ([StructuredState] Cannot simulate a single qubit circuit #156)

Also included some bugfixes of MPSxMPO (I had missed calling _flush() in the newer functions).

Related issues

#151
#156

Checklist

• I have run the tests on a device with GPUs.
• I have performed a self-review of my code.
• I have commented hard-to-understand parts of my code.
• I have made corresponding changes to the public API documentation.
• I have added tests that prove my fix is effective or that my feature works.
• I have updated the changelog with any user-facing changes.

[PabloAndresCQ]

This is now hiding two options that I have decided to deprecate:

• sort_gates that was always applied when calling simulate. When set to True, this is done by _get_sorted_gates (see its docstring). This code is not immediately compatible with classical operations and, although it could be adapted, this is not a priority. I am not convinced it was making a noticeable impact in simulation time for MPS, and for TTN this was probably a bit too naive. The code is not particularly easy to maintain and extend, so I am leaning towards deprecating it by "hiding it" and setting it to False by default.
• use_kahypar used to be a option in Config, but it makes more sense to set it here, since it is the only place where it is used. Moreover, it became clear that further work would be required to include KaHyPar in the default pipeline (due to dependencies), so it was already set to False by default.

[PabloAndresCQ]

The semantics are not agreed upon across devices and emulators. For the time being, I'd rather error out as currently done. The current implementation should be capable of running any circuit that comes from a QASM file (parsed via pytket.qasm) using the standard QASM library. These extra operators on registers come from our internal extension to QASM, hqslib1.inc.

[PabloAndresCQ]

There was a silly bug here. Now fixed.

[jake-arkinstall]

Changes are clearly useful, and tests pass in nix (aside from one in which fidelity is better than expected, which we will cover in a separate PR by adding lower bounds to tests rather than approx checks).