`nidmfsl` run taints processed dataset

[mih]

Running the tool in a dataset creates a bunch of files that are not cleaned up upon exit. Here is an illustrative git status:

% git st
On branch master
Your branch is up to date with 'origin/master'.

Untracked files:
  (use "git add <file>..." to include in what will be committed)

        sub-01/2ndlvl.gfeat/cope1.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope1.feat/tmp_clustmap_01_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope10.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope10.feat/tmp_clustmap_10_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope11.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope11.feat/tmp_clustmap_11_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope12.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope12.feat/tmp_clustmap_12_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope13.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope13.feat/tmp_clustmap_13_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope14.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope14.feat/tmp_clustmap_14_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope15.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope15.feat/tmp_clustmap_15_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope2.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope2.feat/tmp_clustmap_02_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope3.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope3.feat/tmp_clustmap_03_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope4.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope4.feat/tmp_clustmap_04_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope5.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope5.feat/tmp_clustmap_05_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope6.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope6.feat/tmp_clustmap_06_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope7.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope8.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope8.feat/tmp_clustmap_08_T001.nii.gz
        sub-01/2ndlvl.gfeat/cope9.feat/stats/smoothness_v
        sub-01/2ndlvl.gfeat/cope9.feat/tmp_clustmap_09_T001.nii.gz

[cmaumet]

Thanks @mih! Agreed on the smoothness_v file, I'll remove this one. About the tmp_clustmap_*** files could this be related to a failed export. As far as I can see (but obviously could be wrong) they should be deleted if the export runs smoothly.

[cmaumet]

Discussed w/ @mih today: think about the usecase where we don't have write access to the results directory.