Fix failing splicing compute tests and get rid of various code smells

CONTRIBUTING.md

@@ -106,7 +106,10 @@ before the pull request is merged.
 - [ ] Do the new functions have descriptive 
   [numpydoc](https://github.com/numpy/numpy/blob/master/doc/HOWTO_DOCUMENT.rst.txt)
   style docstrings?
-- [ ] If it adds a new plot, is it documented in the gallery?
+- [ ] If it adds a new feature, is it documented as an IPython Notebook in 
+  `examples/`, and is that notebook added to `doc/tutorial.rst`?
+- [ ] If it adds a new plot, is it documented in the gallery, as a `.py` file 
+  in `examples/`?
 - [ ] Is it well formatted? Look at `make pep8` and `make lint` output
 - [ ] Is it documented in the doc/releases/?
 - [ ] Was a spellchecker run on the source code and documentation after

doc/releases/v0.2.6.txt

@@ -6,6 +6,9 @@ This is a patch release, with non-breaking changes from 0.2.5.
 New features
 ~~~~~~~~~~~~
 
+- Add a :py:class:`.data_model.SupplementalData` data type, which allows the
+  user to store any ``pandas.DataFrame`` on the :py:class:`.data_model.Study`
+  object as `study.supplemental`. This is essentially user-driven caching.
 
 Plotting functions
 ~~~~~~~~~~~~~~~~~~
@@ -28,4 +31,4 @@ Miscellaneous
 
 - Streamlined test suite to test fewer things at a time, which shortened the
   test suite from ~20 minutes to ~3 minutes, about 85% time savings.
-- Improved accuracy (fewer false positives) in splicing modality estimation
+- Improved accuracy (fewer false positives) in splicing modality estimation