update snakefile to train model for plmutils instead of rnasamba

[taylorreiter]

PR checklist

• [X] Describe the changes you've made.
• [X] Describe any tests you have conducted to confirm that your changes behave as expected.
• [X] If you've added new software dependencies, make sure that those dependencies are included in the appropriate conda environments.
• [X] If you've added new functionality, make sure that the documentation is updated accordingly.
• [X] If you encountered bugs or features that you won't address, but should be addressed eventually, create new issues for them.

This PR builds a plmutils model using short sequences from ensembl model organisms. It removes the model build for RNAsamba bc we outperform that tool (which this PR also tests...we get 28% accuracy 🎉 while that isn't good, it's better than rnasamba on this data set).

Validation metrics
Auc_roc: 0.82
Accuracy: 0.75
Precision: 0.78
Recall: 0.67
Mcc: 0.50
Num_true_positive: 7343.00
Num_false_positive: 2130.00
Num_true_negative: 10187.00
Num_false_negative: 3690.00
Num_positive: 11033.00
Num_negative: 12317.00
Model saved to 'outputs/models/plmutils/2_model/'

The point of this snakefile is only to document the model build process. I don't expect many people will want to rerun the script (unlike peptigate, which we expect to be run many times).

it also breaks up some comment lines to respect the 100 character limit 🙃