parameterize table names

[bitsofbits]

Closes #21

• Parameterize table names in SQL files
• Create config file for table names

After this is merged, I plan to update the table names, rerun and cut another release.

@pwoods25443 a couple of questions:

• I did not parameterize the table names in the known_fishing queries as those tables looked like they might be stable and there is already one query per year. Is that reasonable?
• Currently some of the data in the other queries is being pulled from pipeline_normalize and some from pipeline_classify_logistic_661b. Do you recall if there is a reason for this, or can I try to consolidate these down to a single table?

[pwoods25443]

@bitsofbits

• I agree that the known fishing queries do not need to use different tables, so the changes in this PR look good
• the pipeline_classify tables are a superset of the pipeline_normalize tables, so we probably could use the same table for everything, but we might change things in the future so that is no longer true. I think we should probably continue to use the normalize table when we are only interested in the raw data and the classify table when we want the processed data