Fix TSP test

[stavros11]

@shangtai this implements the TSP test by comparing the state vector, instead of doing the measurements, thus avoiding the issues with the seed. Please have a look and let me know if you agree.

I also added a copy in the loss function of the QAOA as it was causing the numba backend to fail, most likely due to in-place updates. Perhaps this was the reason we could not get it to work with the measurements either.

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (abstrtsp@b5f351d). Click here to learn what that means.
The diff coverage is n/a.

@@             Coverage Diff              @@
##             abstrtsp      #620   +/-   ##
============================================
  Coverage            ?   100.00%           
============================================
  Files               ?        79           
  Lines               ?     10416           
  Branches            ?         0           
============================================
  Hits                ?     10416           
  Misses              ?         0           
  Partials            ?         0           

Flag	Coverage Δ
unittests	100.00% <0.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b5f351d...0aa0e6c. Read the comment docs.

[scarrazza]

@shangtai could you please have a look and confirm if this solution works for you?

[shangtai]

Hi @stavros11 I think the change is great since we can now avoid the issue with the random number seed.