Refactor metaP aggregation

[kheal]

We've decided to rewrite how the metaproteomics data are being aggregated.

Goals of the script

1. Write to the functional_aggreagation_agg collection
2. Use the API, reduce reliance with direct connection to MongoDB
3. Access the raw protein reports (not just the mongo data) to generate the records
4. Validate and submit data via API

[aclum]

2 & 4 overlaps with #11 so feel free to close or assign to yourself as a subtask.