Issue #1602 - Add get_label_instances to Analysis

[DavidKoleczek]

Description of proposed changes

Implements changes as discussed @bhancock8 in #1602. Added a method to analysis/error_analysis that wraps get_label_buckets functionality. Given a bucket, a NumPy array x of your data, and corresponding y label(s), it will return to you x with only the instances corresponding to that bucket.

Let me know if there are any issues, questions, or suggestions. Thanks!

Related issue(s)

Issue #1602

Checklist

Need help on these? Just ask!

• [x] I have read the CONTRIBUTING document.
• [x] I have updated the documentation accordingly.
• [x] I have added tests to cover my changes.
• [x] I have run tox -e complex and/or tox -e spark if appropriate.
• [x] All new and existing tests passed.

[codecov[bot]]

Codecov Report

Merging #1608 into master will increase coverage by 0.01%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1608      +/-   ##
==========================================
+ Coverage   97.19%   97.21%   +0.01%     
==========================================
  Files          68       68              
  Lines        2137     2151      +14     
  Branches      343      345       +2     
==========================================
+ Hits         2077     2091      +14     
  Misses         31       31              
  Partials       29       29              

Impacted Files	Coverage Δ
snorkel/analysis/__init__.py	100.00% <100.00%> (ø)
snorkel/analysis/error_analysis.py	100.00% <100.00%> (ø)

[DavidKoleczek]

Thanks for the great comments @bhancock8 ! I made the requested changes, let me know if there is anything else.