Networkml

- models have been retrained to fix a warning about invalid results when evaluating a pcap
- some unused code and module has been removed
- upgraded pytest to 4.1.0 and pytest-cov to 2.6.1

- upgraded scikit-learn to 0.20.2
- removed scipy
- cleaned up requirements.txt and setup.py
- fixed issue where redis was throwing error when saving decisions
- fixed error in eval_onelayer that was using nonexistent key
- Make train/eval/test process consistent for all models
- Fixed path error specific to python 3.5 that occurred when processing PCAP files
- PCAP directories can now be used when running model evals

- upgraded pytest to 4.0.2
- upgraded scikit-learn to 0.20.1
- improved README documentation
- upgraded redis to 3.0.1
- added pcap directory support
- re-enabled the behavior model
- includes the trained behavior model
- fixed hardcoded onelayer pickle file in randomforest
- fixed missing labels
- simplified rabbit connection
- replaced deprecated randomized logistic regression with random forest

- upgraded pytest to 3.9.1
- fixed a NoneType error when multiplying
- fixed an issue where the config file wasn't being read properly
- abstracted away the code to read the config file into one place

- lots of cleanup of duplicated code
- upgraded tensorflow to 1.11.0
- upgraded scikit-learn to 0.20.0
- updated the model

- moved a bunch of duplicated code into common utils