Raise exception on nonstandard spaces with modifiers

[tsalo]

Closes #995.

This change raises an exception if users request a nonstandard output space (e.g., bold, asl, anat) with a resolution modifier, since that behavior isn't supported. Currently, at least in ASLPrep and fMRIPrep, if you passed in something like --output-spaces anat:res-01:res-02 you'd just get out one set of anatomical-space derivatives with the native anatomical resolution, which is confusing both because you don't get different resolutions and because you only get one set even though you requested two (res-01 and res-02).

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 71.23%. Comparing base (9d258db) to head (a8fe5a8).
⚠️ Report is 13 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #996      +/-   ##
==========================================
+ Coverage   71.20%   71.23%   +0.03%     
==========================================
  Files          89       89              
  Lines        8578     8588      +10     
  Branches      931      933       +2     
==========================================
+ Hits         6108     6118      +10     
  Misses       2238     2238              
  Partials      232      232              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[tsalo]

This is a solid change for fMRIPrep and ASLPrep but I want to make sure that the other preps don't have special workarounds for it. @oesteban @mnoergaard @mgxd is this change alright for dMRIPrep, PETPrep, and Nibabies?

[oesteban]

Originating PR: nipreps/fmriprep#1604

In the documentation, we state that:

Modifiers are not allowed when providing nonstandard spaces.

So I guess this is actually implementing that part of the contract.

In the future, it'd be fantastic to have a language to define the output resolution in space and time.

[tsalo]

@oesteban I just threw together a basic proposal for a new syntax in #997. Happy to modify it.