fix a bug in conversion from z statistics to p values

[yifan0330]

Closes #748.

Changes proposed in this pull request:

• one-sided right-tail test should take absolute value of z statistics.

[codecov[bot]]

Codecov Report

Base: 88.55% // Head: 88.05% // Decreases project coverage by -0.50% :warning:

Coverage data is based on head (06d235d) compared to base (e269941). Patch coverage: 100.00% of modified lines in pull request are covered.

:exclamation: Current head 06d235d differs from pull request most recent head 89a1dd3. Consider uploading reports for the commit 89a1dd3 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #749      +/-   ##
==========================================
- Coverage   88.55%   88.05%   -0.51%     
==========================================
  Files          38       35       -3     
  Lines        4370     3884     -486     
==========================================
- Hits         3870     3420     -450     
+ Misses        500      464      -36     

Impacted Files	Coverage Δ
nimare/transforms.py	75.36% <100.00%> (ø)
nimare/base.py
nimare/__init__.py
nimare/utils.py

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.

[yifan0330]

Thanks @yifan0330! could you add a test to the test_z_to_p function demonstrating the fix? add something like (-1.959963, "one", 0.975),

https://github.com/neurostuff/NiMARE/blob/13d7b196e95f481f2c6b9cda7411ced942075935/nimare/tests/test_transforms.py#L249-L264

Sorry I just saw your comments! Thanks for the suggestion @jdkent, I've added this test