Small fixes for TWPA calibration routines

[stavros11]

Some issues I realized while trying to use these routines:

1. frequency_power was giving some KeyErrors with qubits missing from the initial_twpa_freq and initial_twpa_power dictionaries. I believe these are because the execution part of the routine is nested in the loop that initializates these dictionaries, while I think it should be a separate loop.
2. It seems that relaxation_time is ignored for both SNR routines as it is not passed in the resonator_spectroscopy call. This ends up using the default relaxation_time from the platform runcard, which is usually much higher than what we need for spectroscopy, thus making execution slower.

Checklist:

• Reviewers confirm new code works as expected.
• Tests are passing.
• Coverage does not decrease.
• Documentation is updated.
• Compatibility with Qibo modules (Please edit this section if the current pull request is not compatible with the following branches).
  • Qibo: master
  • Qibolab: main
  • Qibolab_platforms_qrc: main

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 97.44%. Comparing base (0625d90) to head (79cd1cb).
Report is 16 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #899      +/-   ##
==========================================
- Coverage   97.48%   97.44%   -0.05%     
==========================================
  Files         116      116              
  Lines        8882     8881       -1     
==========================================
- Hits         8659     8654       -5     
- Misses        223      227       +4     

Flag	Coverage Δ
unittests	97.44% <100.00%> (-0.05%)	⬇️

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

Files	Coverage Δ
...out_optimization/twpa_calibration/frequency_SNR.py	89.28% <ø> (ø)
...t_optimization/twpa_calibration/frequency_power.py	100.00% <100.00%> (ø)
...readout_optimization/twpa_calibration/power_SNR.py	89.28% <ø> (ø)

... and 2 files with indirect coverage changes

[andrea-pasquale]

Thanks @stavros11, I agree with all the fixes.
Regarding the relaxation I already had a feeling that it was not used correctly in all protocols (#829).

[Edoardo-Pedicillo]

Suggested change

for qubit in targets:

At this point, I suggest removing this line and leaving the indentation as in main.

[stavros11]

Thanks, both comments are right. I reverted the indentation and removed this in 79cd1cb.

The routine should now work, but we should keep in mind that it is suboptimal and maybe open an issue for this. The acquisition is repeated in multiplexed fashion on all qubits multiple times and in each case we drop the results for all qubits except of one. We should only need one acquisition per TWPA pump, not per qubit. For 5-qubits that share a readout line, which is the case for most of our current chips, we should only need one acquisition, not five.

[Edoardo-Pedicillo]

We have already opened #564 , for sure what we are doing is wrong, but we could collect all the results and find a better way to use them