Add an inflation factor to correct for multiple contrasts in Stouffer's combination test

[JulioAPeraza]

This PR expands Stouffer's combination test to correct for multiple contrasts. Developed by @nicholst.
In this method, we add a correction factor to account for repeated scans (i.e., multiple samples from the same study).

A braking change here that might affect the current implementation of Stouffer's IBMA in NiMARE is the dataset attribute mapping in StoufferCombinationTest. Previously, we mapped the v (study-level variance) to w (weights for weighted Stouffer's). In this PR, I thought of mapping the n (study-level sample sizes) to w and then v to g (array of group labels to determine the inflation factor).

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.07%. Comparing base (bee163b) to head (19fef11).

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #117      +/-   ##
==========================================
+ Coverage   87.79%   88.07%   +0.28%     
==========================================
  Files          13       13              
  Lines         893      914      +21     
==========================================
+ Hits          784      805      +21     
  Misses        109      109              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[jdkent]

I've looked through this, but did not confirm the equations, I think this is fine, so I'm not going to hold this up any longer