Fix for test_add

[Cdebus]

Fixes #164

• Error was result of uncomplete cleanup after restructuring of binary operator generic function
• removal of faulty tensor
• tests on NotImplementedError raise are performed with T_s

[codecov-io]

Codecov Report

Merging #191 into master will decrease coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master    #191      +/-   ##
=========================================
- Coverage   89.61%   89.6%   -0.01%     
=========================================
  Files          36      36              
  Lines        3639    3637       -2     
=========================================
- Hits         3261    3259       -2     
  Misses        378     378

Impacted Files	Coverage Δ
heat/core/tests/test_arithmetics.py	100% <100%> (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 e31ac4f...03a9a79. Read the comment docs.

[coquelin77]

everything passes here with up to 4 processes.

I was looking through the test though and I found a few stylistic things. Such as I didnt see why when you copy the T1 tensor to form T_s you say you want comm=None. I changed that to T1.comm and I also changed the device to T1.device. I also did some pep8 formating in the two lines following.

T_s = ht.tensor(T1._tensor__array, T1.shape, T1.dtype, 0, T1.device, T1.comm)
Ts = ht.ones((2, 2), split=1)
otherType = (2, 2)

[Cdebus]

Well, to be honest, there is no reason^^.... I didn't think about this at the time, I only cared for it to be a split tensor. But it does make a lot more sense to copy the device and communication. Thanks for catching it!