Add initial working multigraph edge validator plus tests

[j6k4m8]

Based upon the work in #106 (thanks @celiibrendan!) and fixes #101.

[codecov-commenter]

Codecov Report

Merging #107 (e827af5) into master (c60ba4e) will increase coverage by 0.41%.
The diff coverage is 97.95%.

❗ Current head e827af5 differs from pull request most recent head 1144b00. Consider uploading reports for the commit 1144b00 to get more accurate results

@@            Coverage Diff             @@
##           master     #107      +/-   ##
==========================================
+ Coverage   88.03%   88.44%   +0.41%     
==========================================
  Files          24       25       +1     
  Lines        1763     1843      +80     
==========================================
+ Hits         1552     1630      +78     
- Misses        211      213       +2     

Impacted Files	Coverage Δ
dotmotif/utils.py	57.14% <ø> (ø)
dotmotif/executors/test_neuprintexecutor.py	33.33% <33.33%> (ø)
dotmotif/tests/test_dm_flags.py	84.61% <87.50%> (ø)
dotmotif/executors/NetworkXExecutor.py	93.02% <95.91%> (+1.35%)	⬆️
dotmotif/__init__.py	95.34% <100.00%> (-0.06%)	⬇️
dotmotif/executors/GrandIsoExecutor.py	80.00% <100.00%> (+0.58%)	⬆️
dotmotif/executors/Neo4jExecutor.py	69.33% <100.00%> (ø)
dotmotif/executors/NeuPrintExecutor.py	41.93% <100.00%> (ø)
dotmotif/executors/test_dm_cypher.py	100.00% <100.00%> (ø)
dotmotif/executors/test_grandisoexecutor.py	100.00% <100.00%> (ø)
... and 7 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 c60ba4e...1144b00. Read the comment docs.

[j6k4m8]

@hpawa and @celiibrendan — I am finding (not unexpectedly) that proper multigraph support is going to take a little more engineering here. For now, I think it's reasonable to use the add-multigraph-support branch (there is GrandIsoExecutor support now, which means that it is quite fast for the types of use cases we've discussed before) while I work on the remaining corner cases.

Here's one such case, to illustrate what I mean:

A -> B [size > 10, size < 10]

Imagine there are two edges between host nodes a and b with size attributes 5 and 20. Right now, the logic doesn't fully support matching this as a success; that is, we only support matching ONE edge per attribute right now.

I'm going to work on building a better system when my brain isn't fried, but I think this is doable. Just... perhaps not tonight :)