call_spew: Wrapper for reading, formatting, and writing SPEW ecosystems
In spew: SPEW Framework for Generating Synthetic Ecosystems

Generates SPEW synthetic ecosystems on the Olympus Computing Cluster

call_spew(input_dir, output_dir, folders = NULL, data_group = "US",
  run_type = "SEQ", sampling_method = "uniform",
  locations_method = "uniform", output_type = "write",
  convert_count = FALSE, vars = list(household = NA, person = NA),
  road_noise = 2e-04, outfile_loc = "", timer = TRUE, verbose = TRUE)

`input_dir`	character specifying input directory
`output_dir`	charcater specifying output directory
`folders`	list specifying sub-directories for each data-source
`data_group`	character, either "US", "ipums", or "none"
`run_type`	Whether to run sequentially in parallel. Default is "SEQ", for a sequential run. If parallel, back-end is either "MPI", "SOCK", or "MC"
`sampling_method`	character indicating the type of sampling method to use, defaults to "uniform". Can also be "ipf" with appropriate marginal data.
`locations_method`	character indicating the type of location sampling to use, defaults to "uniform", can also be "roads".
`output_type`	Default is "console" if we want to resulting population as an R variable. Alternative is "write", which is used on Olympus for writing out .csv files of the population
`convert_count`	logical meant to indicate if we are going to convert population totals from people to household counts. Default: FALSE, assumes the population is the total number of households
`vars`	list with two elements: household and person. This specifies which variables to include in the corresponding household and person PUMS data-set
`road_noise`	Noise added to households during road-based sampling
`outfile_loc`	Defaults to "", so we print out the parallel run information. Only set to "/dev/null" for internal testing putposes.
`timer`	logical indicating we want to time the run
`verbose`	logical indicating we want to print output during the run. Default is FALSE.#'