[FIX] Remove father-level for meg filetypes other than BTi/4D data

[teonbrooks]

before the document migration, we were wrapping up discussion on removing the father-level of folders for all MEG formats except for BTi/4D. The premise of the father-level directory was to compensate for the fact that BTi data is written without file extension. Its data files are generated with the assumption that they all belong to the same folder.

we proposed adding an additional folder level for this data format. it was extended to a few other file formats but we have realized that this is unnecessary breaks the general pattern. we then proposed to remove the father-level for the file types affected by the superfluous folder level with the exception of BTi where the folder isn't superfluous but necessary.

[chrisgorgo]

This has been brought up by @teonbrooks before (https://groups.google.com/d/msgid/bids-discussion/CAGdVzab%3DSufiv4Qs4CWwSLTDXs0C6sfMVJG3YH4Vz0VT%2BdyaOw%40mail.gmail.com?utm_medium=email&utm_source=footer). It would be great to get @guiomar and @robertoostenveld opinion.

[teonbrooks]

here's the discussion I had with @robertoostenveld from the google doc.

[sappelhoff]

@robertoostenveld @monkeyman192 could you find some time to review this?

It looks good to me but I'd like some MEG expert input

[robertoostenveld]

I think it would be good to make the initial table more consistent, i.e. in the definition (2nd column) only list the timeseries data (plus header and/or events), or to list the extension of all files that are mentioned in the manufacturer specific paragraph. Either one is fine with me.

for KIT .mrk is listed in the table. I think it is for head markers, not events. But I am not sure. If all files are in the table, then also the others should be completed in the table.

For CTF the .shape and .shape_info are completely missing, they are supported by the validator.

For KRISS the .chn and .trg are listed in the paragraph, not in the table.

Aalto MEG should refer to "Neuromag/Elekta/MEGIN" format (now MEGIN is not part of it).

But if you want to keep the PR small, I am fine and approve these changes.

[monkeyman192]

I think having all the raw files in the same folder is a good idea (I think I brought it up initially in the google docs.)
The changes look good and I think that the changes @robertoostenveld mentioned are probably better done in a separate PR to keep the issues separate.
I just have the one suggested change but LGTM otherwise :)

[monkeyman192]

You should add that .sqd is a valid marker file extension also

[sappelhoff]

Thanks for the review @monkeyman192 I am merging this now and will ask you to incorporate your review comment in your own PR #62 because you are editing the same file there and both PRs are connected (slightly) :-)

@teonbrooks will probably be happy to have this closed after 5 months 😆

Thanks everybody

[jasmainak]

Does this need any updates in the validator?

[sappelhoff]

Does this need any updates in the validator?

see mne-tools/mne-bids#187 (comment)

[teonbrooks]

awesome! @sappelhoff, you're correct, i am pleased :)