Refactor `gscontrol` module

[tsalo]

Closes #1085 (but perhaps not optimally?)

Changes proposed in this pull request:

• Add type hints to gscontrol.gscontrol_raw and gscontrol.minimum_image_regression.
• Use the same variable names from metric calculation within these two functions as well. E.g., data_cat instead of catd.
• Require function calls to use keyword arguments.
• Add verbose parameter to gscontrol.minimum_image_regression. Only write out the accepted MIR outputs if verbose=True.
  • Note that these are only composed of non-low-variance components at the moment.
• In minimum_image_regression, use classification tags to split up accepted components into BOLD-like and low-variance groups, to match the old "accepted" and "ignored" classifications.

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 95.71429% with 3 lines in your changes are missing coverage. Please review.

Project coverage is 89.87%. Comparing base (0f6cbe1) to head (c491cc3).

Files	Patch %	Lines
tedana/gscontrol.py	94.91%	2 Missing and 1 partial :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #1086 +/- ## ========================================== - Coverage 89.91% 89.87% -0.05% ========================================== Files 26 26 Lines 3621 3634 +13 Branches 629 632 +3 ========================================== + Hits 3256 3266 +10 - Misses 214 216 +2 - Partials 151 152 +1 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.