DB: Check for duplicate incidents with case tag

[rxun]

Summary
There is a potential for duplicate incidents in multiple datasets (Massage Parlor Trafficking, Prostitution Arrests or Stings, Human Trafficking Arrests) -- during preprocessing, we should check for whether or not there are duplicates.

Solution

• Every incident should have a unique casetag
• (???) Prioritize "Prostitution arrests or stings" (since these will indicate whether victim and trafficker were both arrested), then Massage Parlor Trafficking, and finally, Human Trafficking Arrests