RE: CSV files with Byte Over Marks

Yuck.
The bad news is that your system is configured as though it were the year 1990. The good news is that you can change that. I changed mine at a time when I applied PTFs and needed to do an IPL anyway, but there are definitely people who changed it in the middle of the day.
CHGSYSVAL QCCSID(37)

I'm not surprised. We haven't had a system admin since 2008. I'm pretty much it on our IBM i. I'm the programmer, system admin, and system operator. I'll see about changing the character ID. Maybe when we upgrade to V7R3, then that is when we'll do it.

The best news is that you can change just /your job/ immediately and test again!
CHGJOB CCSID(37) - Use 37 if you are in the US and have an English language installed.
Test the program again.

What character ID does the file need to be on the IFS? Do I changed it back to 1252 as it was when I exported it from ADP?

If that works (and I suspect it will), you can make this process run correctly without changing the system CCSID by doing a SBMJOB ...
CCSID(37). Or using a user profile whose CCSID is 37.

Note for the archives: 37 should only be used if you want US English EBCDIC.

The file is on the IFS. It currently says " 1208", because that is what I changed it to using the CHGATR command as suggested earlier. What it was before was 1252. This format has the BOM which is what I'm trying to get rid of.

I'd say that the BOM isn't the issue: the CCSID(65535) probably is.
I'd also say that you probably want to keep the BOM, but it seems quick to test.

When I run the test using CCSID 37 and the file on the IFS is as I exported it from ADP I'm still getting the funky characters as part of the first column I'm reading.

Normally, files destined for Windows UTF-8 has the BOM, and there's definitely a BOM in this data. So I'd call it UTF-8. That said, UTF-8 is a variable length encoding, where one 'character' can be encoded with one, two, three, or four 8-bit bytes. By rule, the first 128 'characters' are mapped to the same
8-bit encoding that ASCII 819 is mapped to. This is for compatibility with older software.
https://en.wikipedia.org/wiki/UTF-8