create local data store for database data

[jmlondon]

Many data components of this analysis rely on a pull from an internal PostgreSQL database. This has two implications:

1. the data on the database can change/update and impact downstream analysis and results along with future reproducibility
2. there's no option to version control the data

To resolve this, the plan is to create local data stores within the repository as either *.csv or *.parquet files

[jmlondon]

going to explore managing this data store with the piggyback package

https://cran.r-project.org/web/packages/piggyback/vignettes/intro.html

[jmlondon]

my plan for this is to manage 3 repos related to this analysis/manuscript:

1. berchukseals-haulout (this repo) will house the manuscript and core analysis/model fitting data. data will be pulled from the (new) berchukseals-data repo
2. berchukseals-fits will house the final model fits as releases that can be easily downloaded for us in other analyses
3. berchukseals-data will house the raw data and data processing scripts that involve interaction with the internal NOAA/PEP database. data products will be available as releases