Earlier quoted context omitted.
100% right. The challenge is it's not cheap, and someone has to pay for that and it's either government (probably should be, but we're waiting...) or the private sector, and it seems to be a hard revenue model to make work.
You first identify "most likely to be contaminated foods" whether that is a type, or sales method, or handling, etc. Then you require that companies test and report some metric associated with the issue, just like they report calories and ingredients. Then you require and fund the government to spot-check the reported amounts. Eventually you could even require minimums.
You can't fit a model without data. Consumer Reports did some random spot tests and found phtalates where we didn't expect them. That means we need more broad-based data to be able to predict what is and isn't likely to be contaminated.