Fix transpiler bugs about moved measurements and duplicated gates

[Simone-Bordoni]

Solve bugs in #1363 and #1318

Checklist:

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

[Simone-Bordoni]

@BrunoLiegiBastonLiegi I have solved the bug reported in #1363. I will try to rapidly fix the Issue in #1318, in the meantime you can start the review of this part (I was not reardering the initial qubit map and this created some issues with the random placer). I have also added the seed for the random placer

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.84%. Comparing base (88825fa) to head (d78b626).

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1369   +/-   ##
=======================================
  Coverage   99.84%   99.84%           
=======================================
  Files          75       75           
  Lines       10799    10807    +8     
=======================================
+ Hits        10782    10790    +8     
  Misses         17       17           

Flag	Coverage Δ
unittests	99.84% <100.00%> (+<0.01%)	⬆️

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.

[renatomello]

@Simone-Bordoni Can the PR name be more descriptive?

[BrunoLiegiBastonLiegi]

Thanks @Simone-Bordoni I tested this with #1106 and it seems to be working

[renatomello]

A couple of other things:

1. the function translate_gate in unroller.py receives the variable gate as an input, but redefines what gate is in a loop.
2. the Custom.__call__ method in placer.py receives the variable map as an input, but map is a python built-in function. This may lead to issues, so this variable should be renamed.

[andrea-pasquale]

Thanks @Simone-Bordoni.
I can confirm that this PR fixes #1318.
While testing the state_tomography in Qibocal, I noticed that even without this PR is now failing with this error:

File "/home/andreapasquale/qibocal/src/qibocal/protocols/state_tomography.py", line 114, in _acquisition
    _, results = execute_transpiled_circuit(
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/andreapasquale/qibocal/src/qibocal/auto/transpile.py", line 60, in execute_transpiled_circuit
    transpiled_circ, _ = transpiler(new_circuit)
                         ^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/andreapasquale/qibo/src/qibo/transpiler/pipeline.py", line 265, in __call__
    return circuit, final_layout
                    ^^^^^^^^^^^^
UnboundLocalError: cannot access local variable 'final_layout' where it is not associated with a value

I believe that the error was introduced by 8d29397. It should be related to the fact that we are returning final_layout even if it is defined only in a if statement. Probably in qibocal we fall in one of the cases where final_layout is not defined.
I believe that this was related to the changes that @Edoardo-Pedicillo introduced in #1301.
Given that the error by itself is not related to this PR feel free to address the problem here or in a separate PR and I will open an issue pointing to this comment.