create -d DSPATH -c procedure SUBDSPATH leaves DSPATH dirty

[yarikoptic]

I guess either save for adding a new subdataset comes too early (before calling the cfg_ procedure) or we need a second save to store adjusted (by the procedure) state.

$> datalad create /tmp/testds && datalad create -d /tmp/testds -c text2git subds  && datalad -C /tmp/testds status
[INFO   ] Creating a new annex repo at /tmp/testds 
create(ok): /tmp/testds (dataset)                                                                               
[INFO   ] Creating a new annex repo at /tmp/testds/subds 
create(ok): subds (dataset)                                                                                     
[INFO   ] Running procedure cfg_text2git 
[INFO   ] == Command start (output follows) ===== 
[INFO   ] == Command exit (modification check follows) ===== 
action summary:
  add (ok: 2)
  create (ok: 1)
  save (ok: 1)
 modified: subds (dataset)

DataLad 0.12.0rc4-327-g6307824f2
NB Ran into while testing a use case workflow using reproman run.

[yarikoptic]

or may be it is just due to the "feature" of not saving updated state of the subds if there is no explicit --dataset?

$> datalad save -m "Due to https://github.com/datalad/datalad/issues/3591" data/mriqc

$> git status                                                                        
On branch master
Changes not staged for commit:
  (use "git add <file>..." to update what will be committed)
  (use "git checkout -- <file>..." to discard changes in working directory)

	modified:   data/mriqc (new commits)

no changes added to commit (use "git add" and/or "git commit -a")

$> datalad save -m "Due to https://github.com/datalad/datalad/issues/3591" -d . data/mriqc
add(ok): data/mriqc (file)
save(ok): . (dataset)
action summary:
  add (ok: 1)
  save (ok: 1)

[kyleam]

I think it might just be a matter of order. I'll take a closer look.