Remove 'deepencoding' architecture

[VarIr]

Initially, the working title of the main deep network architecture used in deepnog was deepencoding.
For the paper, we switched to calling this DeepNOG, similar to the program itself.
This PR removes deepencoding altogether in order to avoid confusion about naming.

[codecov]

Codecov Report

Merging #32 (21cd493) into master (d62d350) will decrease coverage by 0.19%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master      #32      +/-   ##
==========================================
- Coverage   96.92%   96.72%   -0.20%     
==========================================
  Files          37       36       -1     
  Lines        2112     2045      -67     
==========================================
- Hits         2047     1978      -69     
- Misses         65       67       +2     

Impacted Files	Coverage Δ
deepnog/client/client.py	96.93% <ø> (-1.23%)	⬇️
deepnog/data/dataset.py	97.81% <ø> (ø)
deepnog/models/deepnog.py	96.92% <ø> (ø)
deepnog/tests/tests/test_utils.py	100.00% <ø> (ø)
deepnog/utils/io_utils.py	100.00% <ø> (ø)
deepnog/utils/network.py	91.22% <ø> (-3.51%)	⬇️
deepnog/utils/tests/test_imports.py	100.00% <ø> (ø)
deepnog/client/tests/test_cli.py	99.37% <100.00%> (ø)
deepnog/learning/tests/test_inference.py	100.00% <100.00%> (ø)
deepnog/models/__init__.py	100.00% <100.00%> (ø)
... and 5 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 b23bee4...21cd493. Read the comment docs.

[VarIr]

lgtmyself two days later