DOC: Link out to BIDS Stats Models docs instead of maintaining our own

[Shotgunosine]

We probably don't want to duplicate the stats models documentation in the fintlins docs. Closes #377

[Shotgunosine]

@effigies and @adelavega do y'all have thoughts on this, #375 and #374?

[effigies]

Do we actually cover everything in here, over there? If so, I think dropping model.rst altogether in favor of consistently linking to https://bids-standard.github.io/stats-models makes sense.

[adelavega]

I think we do except the outputs section, perhaps we can leave that in? Otherwise LGTM

[codecov-commenter]

Codecov Report

Merging #378 (47265dd) into dev (372e6d2) will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##              dev     #378   +/-   ##
=======================================
  Coverage   64.42%   64.42%           
=======================================
  Files          23       23           
  Lines        1816     1816           
  Branches      343      343           
=======================================
  Hits         1170     1170           
  Misses        564      564           
  Partials       82       82           

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 372e6d2...47265dd. Read the comment docs.

[Shotgunosine]

@adelavega just getting back to this now. Which outputs section did you mean? Just this sentence? "The output of this level will again be a statistical map for the contrast a_vs_b, but summarized across the whole group."

@effigies, sounds good.

[adelavega]

I think that's all I meant but it doesn't seem very important now. I'm no help w/ the rst/md thing.

Otherwise still LGTM

[effigies]

@Shotgunosine Pushed a fix. Thanks for catching that.

[effigies]

Failures unrelated. Diagnosing separately.