fix: Calculate bold mask and dummy scans in transform-only runs

[effigies]

This builds on #3427 to add a couple final things to the transform-only workflow:

fmriprep $BIDS $OUT participant --derivatives fmriprep=$MINIMAL

BOLD masks were not getting generated from the boldref and dummy scans were not being estimated.

This PR factors out the validation and dummy scan detection from init_raw_boldref_wf and uses the full workflow in fit mode and the subworkflow in transform mode.

Will rebase after #3427 is merged. Can compare now with: c28a615...6c78e54

[codecov]

Codecov Report

Attention: Patch coverage is 93.75000% with 1 line in your changes missing coverage. Please review.

Project coverage is 72.99%. Comparing base (8c58b70) to head (fdbf0f1).

Files with missing lines	Patch %	Lines
fmriprep/workflows/bold/reference.py	90.90%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3428      +/-   ##
==========================================
+ Coverage   72.92%   72.99%   +0.06%     
==========================================
  Files          57       57              
  Lines        4358     4369      +11     
  Branches      547      548       +1     
==========================================
+ Hits         3178     3189      +11     
  Misses       1064     1064              
  Partials      116      116              

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