Refine viz pattern

[padilla410]

This PR completes the mapping portion of the targets pipeline for the NAWQA Pesticide National Synthesis Project.

Building the pipeline creates two major families of outputs for each of the 2019 pesticides of interest:

1. GeoTIFFs for high and low pesticide estimates
2. Map pngs that roughly match the existing maps (example).

There is probably some refinement of the final images that will happen, but I intend to hold off until we meet with the WSC folks.

With no pre-existing targets, I expect this pipeline to take ~80 min to run.

[cnell-usgs]

Looks really nice. The structure of this made sense to me based on my understanding of the data, and what the moving parts are. I was able to build the entire pipeline in 160 mins:

The longest part of the build was generating the plots, but that seems pretty typical of plotting rasters with ggplot

In your create_pest_raster function you can clean up the commented out code creating cellkg since it is used a couple lines later.

I also might combine a few steps in create_pest_raster that are applied to cellkg rather than writing over the same-named object.

cellkg <- as.data.frame(us_raster) %>%
    rename(ID = 1) %>%
    left_join(dbf_data) %>% 
    dplyr::pull({{ cellkg_column }})

[padilla410]

First, thank you for a great set of comments/critiques. Second, because of some timeline stuff, I am going address comments that seems quick and easy within this PR and anything more time consuming will have an issue opened for the next round of work.