fix: disable_progressbar in compute_availabilitymatrix

[Irieo]

Changes proposed in this Pull Request

The function compute_availabilitymatrix() includes an option to disable the progress bar (disable_progressbar). However, for non-parallel computation path (i.e., when nprocesses is None, the code uses tqdm directly, w/o passing the disable_progressbar check. This PR fixes it. User is informed if progress bar is disabled.

Why bother: it seems that progress bar makes computation much slower. For a test case (UA, solar availability matrix, local computation, nprocesses is None) disabling progress bar speeds the process from 2000 seconds to 760 seconds. Similar behavior was observed for other atlite processes, such as build_renewable_profiles (rumor evidence).

Closes #pypsa-eur/1090

Checklist

• [ ] Code changes are sufficiently documented; i.e. new functions contain docstrings and further explanations may be given in doc.
• [ ] A note for the release notes doc/release_notes.rst of the upcoming release is included.
• [ ] I consent to the release of this PR's code under the MIT license.