Update Documentation for wire_names and _Global

[csookim]

All changes are related to variable names, docstrings, and documentation. No changes to the logic.

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 ✅

Project coverage is 99.68%. Comparing base (bbf0ccf) to head (9ef4b40).
Report is 68 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1531   +/-   ##
=======================================
  Coverage   99.68%   99.68%           
=======================================
  Files          80       80           
  Lines       11585    11586    +1     
=======================================
+ Hits        11549    11550    +1     
  Misses         36       36           

Flag	Coverage Δ
unittests	99.68% <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.

[alecandido]

In general, the PR seems alright, and I just have a relevant note about a single point.

However, if I'm correct, this is still partially lacking the explanation for a question that we already received, i.e. the role of wire_names in the qubit selection.
With this, I mean that we should clarify the following points:

1. wire_names is absolutely not limiting the qubits used for placing or routing, if these non-trivial passes are included in the transpiler - thus, using wire_names as input for a non-trivial placer/routing is only possibly helpful to improve the transpiler solution, but it won't prevent the usage of any qubit in the platform
2. wire_names are instead exactly honored by the Qibolab backend, when used directly (no transpilation), leading to compilation errors if the connectivity is insufficient
3. padding should not be required at any stage, since the information that used to provide is now fully replaced by the usage of wire_names and the connectivity, to the point that we should even be able to drop the Preprocessing (though this should be confirmed)