Expectation values as native backend arrays

[BrunoLiegiBastonLiegi]

This updates the expectation values calculation, that are now computed and returned as native arrays of the backend used. This, in turns, begs for updating some of the error mitigation methods accordingly. In particular, as torch was having problems with the scipy.curve_fit I had to implement a custom _curve_fit based on the torch.optim.LBFGS optimizer.

Checklist:

• Reviewers confirm new code works as expected.
• Tests are passing.
• Coverage does not decrease.
• Documentation is updated.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Please upload report for BASE (master@325190e). Learn more about missing BASE report.
Report is 18 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff            @@
##             master    #1382   +/-   ##
=========================================
  Coverage          ?   97.11%           
=========================================
  Files             ?       81           
  Lines             ?    11699           
  Branches          ?        0           
=========================================
  Hits              ?    11362           
  Misses            ?      337           
  Partials          ?        0           

Flag	Coverage Δ
unittests	97.11% <100.00%> (?)

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

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[MatteoRobbiati]

Thanks! I just left a comment about the optimizer choice.