Integrate lightgbm feature contribution

[shengwangsw]

The pull request integrates LightGBM predict feature contributions functionality. In addition, code refactoring.

[codecov[bot]]

Codecov Report

Merging #89 (122734b) into master (886bc04) will decrease coverage by 0.02%. The diff coverage is 80.39%.

@@             Coverage Diff              @@
##             master      #89      +/-   ##
============================================
- Coverage     79.37%   79.34%   -0.03%     
- Complexity      399      406       +7     
============================================
  Files            41       41              
  Lines          1401     1438      +37     
  Branches        128      135       +7     
============================================
+ Hits           1112     1141      +29     
- Misses          219      223       +4     
- Partials         70       74       +4     

Impacted Files	Coverage Δ
...enml/provider/lightgbm/LightGBMDescriptorUtil.java	96.96% <ø> (ø)
...feedzai/openml/provider/lightgbm/LightGBMSWIG.java	79.68% <75.00%> (-1.57%)	:arrow_down:
...eedzai/openml/provider/lightgbm/SWIGResources.java	82.65% <78.94%> (-0.90%)	:arrow_down:
...openml/provider/lightgbm/LightGBMModelCreator.java	87.23% <83.33%> (+0.13%)	:arrow_up:
...er/lightgbm/LightGBMBinaryClassificationModel.java	60.86% <88.88%> (+1.77%)	:arrow_up:
...tgbm/LightGBMBinaryClassificationModelTrainer.java	86.86% <100.00%> (ø)

---

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 886bc04...122734b. Read the comment docs.

[shengwangsw]

addressed comment and rebased against master

[ghost]

Approved. Btw, did you create a tech-debt issue to later rethink the design to avoid the locks? And also to avoid having to repeat the computations of predictions and explanations?