enh: add boldref / sbref to source metadata

[mgxd]

Closes #3498

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 73.73%. Comparing base (089b494) to head (10c20b9).
⚠️ Report is 12 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3532      +/-   ##
==========================================
+ Coverage   73.72%   73.73%   +0.01%     
==========================================
  Files          60       60              
  Lines        4757     4756       -1     
  Branches      615      615              
==========================================
  Hits         3507     3507              
+ Misses       1109     1108       -1     
  Partials      141      141              

☔ 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.

[mgxd]

Looking at the boldref sources for one of the CI runs.. the idea behind this change was to make it easier to trace whether an sbref or the hmc boldref (which is a derivative of the bold series itself) was used to create it. I don't think we should include working directory files in the sources, so we could consider a few options:

• Backtracking to the first 'source' file (raw / derivative) and pass that along as a source
• Saving additional outputs

[effigies]

Just to pull the contents:

sub-02_task-cuedSGT_run-01_desc-coreg_boldref.json

{
  "Sources": [
    "/scratch/fmriprep_25_2_wf/sub_02_wf/bold_task_cuedSGT_run_01_echo_1_wf/bold_fit_wf/hmc_boldref_wf/gen_avg/sub-02_task-cuedSGT_run-01_echo-1_bold_valid_average.nii.gz",
    "bids::sub-02/func/sub-02_task-cuedSGT_run-01_from-boldref_to-auto00000_mode-image_desc-fmap_xfm.txt",
    "/scratch/fmriprep_25_2_wf/sub_02_wf/fmap_preproc_wf/wf_auto_00000/bs_filter/fmap_syn0Warp_Hz_trans_coeff000.nii.gz"
  ]
}

[effigies]

I think this just shows that we're not using the outputs of a datasink in our buffers. I'll fix that.

[effigies]

sub-100185_task-machinegame_run-01_desc-coreg_boldref.json:

{
  "Sources": [
    "bids:raw:sub-100185/func/sub-100185_task-machinegame_run-01_sbref.nii.gz",
    "bids::sub-100185/func/sub-100185_task-machinegame_run-01_from-boldref_to-auto00000_mode-image_desc-fmap_xfm.txt",
    "bids::sub-100185/fmap/sub-100185_fmapid-auto00000_desc-coeff_fieldmap.nii.gz"
  ]
}

sub-100185_task-machinegame_run-01_from-boldref_to-T1w_mode-image_desc-coreg_xfm.json

{
  "GeneratedBy": [
    {
      "Name": "flirt",
      "Version": "<unknown>"
    }
  ],
  "Sources": [
    "bids::sub-100185/func/sub-100185_task-machinegame_run-01_desc-coreg_boldref.nii.gz"
  ]
}

sub-100185_task-machinegame_run-01_from-boldref_to-auto00000_mode-image_desc-fmap_xfm.json

{
  "Sources": [
    "bids:raw:sub-100185/func/sub-100185_task-machinegame_run-01_sbref.nii.gz",
    "bids::sub-100185/fmap/sub-100185_fmapid-auto00000_desc-magnitude_fieldmap.nii.gz"
  ]
}

sub-100185_task-machinegame_run-02_desc-coreg_boldref.json

{
  "Sources": [
    "bids:raw:sub-100185/func/sub-100185_task-machinegame_run-02_sbref.nii.gz",
    "bids::sub-100185/func/sub-100185_task-machinegame_run-02_from-boldref_to-auto00000_mode-image_desc-fmap_xfm.txt",
    "bids::sub-100185/fmap/sub-100185_fmapid-auto00000_desc-coeff_fieldmap.nii.gz"
  ]
}

sub-02_task-cuedSGT_run-01_desc-coreg_boldref.json

{
  "Sources": [
    "bids::sub-02/func/sub-02_task-cuedSGT_run-01_desc-hmc_boldref.nii.gz",
    "bids::sub-02/func/sub-02_task-cuedSGT_run-01_from-boldref_to-auto00000_mode-image_desc-fmap_xfm.txt",
    "bids::sub-02/fmap/sub-02_run-01_fmapid-auto00000_desc-coeff_fieldmap.nii.gz"
  ]
}

This could definitely be fleshed out more, but we're getting fieldmap and boldref/sbref sources from the input/output directories. I would suggest that we merge nipreps/sdcflows#506 and this and open an issue with "ideal" source lists that can be worked toward piecemeal.

After nipreps/sdcflows#506 is merged, I'll rebase this and relock the pixi environment.

[effigies]

I think this is good to go. Please merge if you're happy with it.

[mgxd]

looks great - thanks for the assist