Results of a contains query should be pre-sorted by levenshtein distance
Or at the very least, we should have an option to request them be returned sorted as such.
Scenario:
We have a large table that contains many company names, and a front-end interface that lets users type part of the company name to perform a look-up to find their ID.
However, if one company has a short name that is made up of letters that commonly appear in many other company names (for example, imagine a company called "Inc inc."), there is no way for us to make sure that a user typing "Inc" into the "company name contains" search query, would see the "Inc inc" at the top of the result list.
And worse, if we're limiting the number of results returned by the contains query, it's likely that "Inc inc." won't even be in the result set at all.
So we're forced to increase the size of our query limit and then run a levenshtein function on all the results on the frontend instead of having the data returned pre-sorted.