Update pipeline for SCT's master branch version

[NadiaBlostein]

Updates:

• SCT packages up-to-date

• incorporated the new version of the get_centerline function, created to facilitate the conversion from pixel to physical space according to @rohanbanerjee and @benjamindeleener’s original diagnosis

• updated preprocessing code to work with the latest version of SCT, i.e. upcoming 6.0 release

• updated according to BIDS convention – agreed upon with @rohanbanerjee during previous discussion

• instead of using the default average_vert_length dictionary, this version a) creates its own average_vert_length dictionary that is adapted to your specific dataset and b) makes the template go as low as the lowest disc at a quantile of your choice in your dataset (e.g. 0) (replacing information for subjects for whom this is missing with the new average_vert_length dictionary)

• tackled issue #17

• made clearer README and added instructions / code on how people should segment the SC and discs.

[NadiaBlostein]

Hi Joshua!!! Thank you very much for the thorough feedback and advice; I will keep those tips in mind moving forward. I am not sure whether there is any open data that can be used as an example but I will ask (and make sure to remain consistent with nomenclature; sometimes when you see philadelphia-pediatric, it's because I missed that part and did not make it generic. Thanks again!