Bug/490 gnb testing

[ClaudiaComito]

Description

This PR fixes assorted bugs that led test_gaussiannb to fail on 7 nodes. The problems were related to how heat functions/methods deal with locally empty tensors.

Issue/s resolved: #490

Changes proposed:

• ht.nonzero() now returns dndarray.DNDarray() instead of factories.array(), as the latter gets stuck when one of the nodes has no data (see bug distributed array from scalar #492 - which is NOT solved)
• dndarray.getitem in distributed mode now returns the correct result both with 0-dim input (a scalar), and with an empty tensor as input.

Type of change

Remove irrelevant options:

• Bug fix (non-breaking change which fixes an issue)

Due Diligence

• All split configurations tested
• Multiple dtypes tested in relevant functions
• Documentation updated (if needed)
• Updated changelog.md under the title "Pending Additions"

Does this change modify the behaviour of other functions? If so, which?

no

[codecov]

Codecov Report

Merging #519 into master will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #519      +/-   ##
==========================================
+ Coverage   96.57%   96.59%   +0.01%     
==========================================
  Files          68       68              
  Lines       14084    14089       +5     
==========================================
+ Hits        13602    13609       +7     
+ Misses        482      480       -2     

Impacted Files	Coverage Δ
heat/core/dndarray.py	96.79% <100.00%> (+0.03%)	⬆️
heat/core/indexing.py	100.00% <100.00%> (ø)
heat/naive_bayes/tests/test_gaussiannb.py	96.03% <100.00%> (-0.07%)	⬇️
heat/core/statistics.py	97.82% <0.00%> (+0.43%)	⬆️

---

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 48432ed...ce264e9. Read the comment docs.