Consider filling in county_fips codes when we have a county and state name

[joshzarrabi]

Is your feature request related to a problem? Please describe.

The COVID dataset has a few rows with counties that don't have a fips code, but do have the state name and county name.

Right now we just toss these out, but in theory we should be able to look up the fips code based on the county and state name and then add it to our dataset. See the screen shot for an example.

Describe the solution you'd like We would look up the county fips code from a public dataset on big query, and then merge that fips code into the dataset.

Then, we would have to add the numbers in those rows to any currently existing rows with the same county_fips.

Additional context We should probably ask the CDC people about this to make sure it is legit.