Kva back to transformer

[NicolasGensollen]

Attempt to address the KVA issues described in #160

@tarekelgindy, this PR impacts basically all parsers so we probably should be very careful and test it (smartd-ds and others) before eventually merging.

NOTES:

• I didn't update the DEW reader since it is not working anyway
• I did my best for the Gridlab-D parsers and the Ephasor writer but I'm not familiar with them, so double checking seems like a good idea

[tarekelgindy]

I'd like to merge some of the other branches into master before we merge this one. The changes look good but there are some new modifications to the regulator writer in CYME (for instance) that should be merged first.

[codecov-io]

Codecov Report

Merging #172 into master will decrease coverage by 0.02%.
The diff coverage is 55.55%.

@@            Coverage Diff             @@
##           master     #172      +/-   ##
==========================================
- Coverage   43.96%   43.94%   -0.03%     
==========================================
  Files          74       74              
  Lines       16473    16383      -90     
==========================================
- Hits         7243     7200      -43     
+ Misses       9230     9183      -47

Impacted Files	Coverage Δ
ditto/models/winding.py	100% <ø> (ø)	⬆️
ditto/writers/json/write.py	93.47% <ø> (ø)	⬆️
ditto/readers/json/read.py	88.88% <ø> (ø)	⬆️
ditto/readers/synergi/read.py	0% <0%> (ø)	⬆️
ditto/readers/opendss/read.py	60.56% <100%> (+0.26%)	⬆️
ditto/writers/ephasor/write.py	90.41% <100%> (ø)	⬆️
ditto/models/regulator.py	100% <100%> (ø)	⬆️
ditto/writers/gridlabd/write.py	84.98% <100%> (ø)	⬆️
ditto/models/powertransformer.py	100% <100%> (ø)	⬆️
ditto/readers/cyme/read.py	62.13% <16.66%> (+0.26%)	⬆️
... and 4 more

---

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 4b7e894...248552a. Read the comment docs.