Flagged Runs & Output Routing

Use a task‑level config switch to control how flagged runs write files and whether they propagate to final outputs.

Current Behavior (default)

save_raw_to_set / save_epochs_to_set write into directories named FLAGGED_<stage> when the run is flagged.
During _entrypoint, post‑completion data is saved with flagged set to the task’s flagged status; final outputs are then skipped if the run is flagged.
copy_final_files ignores any post‑completion directories whose names start with FLAGGED, so flagged runs never appear in the final_files folder.

Introduce a boolean task config move_flagged_files (default: True).

When True (default): flagged runs are quarantined into FLAGGED_* directories and do not produce final files.
When False: flagged runs keep normal directory names and still produce final files. The run remains “flagged” for your tracking, but routing is not altered.

config = {
    # ... your existing settings ...
    "move_flagged_files": False,  # keep normal names and still produce final files
}

With move_flagged_files: True (default):

output/
  FLAGGED_raw_set/
  FLAGGED_epochs_set/
final_files/
  (no files copied from flagged run)

With move_flagged_files: False:

output/
  raw_set/
  epochs_set/
final_files/
  (files copied as usual)

Leave enabled for production QC to quarantine suspect outputs.
Consider disabling temporarily during development or controlled analyses where you must collect outputs regardless of flags.

See Advanced → Task Variables for defining additional task‑specific controls: /advanced/task-variables