Improve file path management when data already downloaded

[brunj7]

Originally, we decided to stop the execution of the download function when the data set had already been downloaded. However, this design is not convenient when using R markdown, especially when you want to knit a document.

Fix: In addition to the warning message, we should also return the path to the existing data. This should allow the code to run even when the dataset has already been downloaded

[mbjones]

Julien -- you might consider using contentid as a caching mechanism, which will seamlessly allow your rmarkdowns to knit from either the cache or the original URL source. See: https://learning.nceas.ucsb.edu/2021-11-delta/session-3-programmatic-metadata-and-data-access.html for some examples.

[brunj7]

Ooh that looks promising, thanks @mbjones ! Let me think about how I can best integrate this into the current workflow

[njlyon0]

Pivot

Currently "solved" by adding _copy# to the end of the created folder if the data already exist. Need to instead provide an option to the user about whether to (A) skip the download if the data already exist or (B) download a copy in the way the function currently behaves.

Julien and I feel the better default behavior is to skip the download because this lets uses of the function in Quarto/Rmarkdown contexts work on the same data without editing the code chunks to fight a new argument.