Feat - Use frecency to rank search results

[PowerUser64]

One of the biggest areas of possible improvements that I see for Kvaesito to make is learning from search results and what the user clicks on to personalize and more accurately rank suggestions. To fill this in gap, I suggest that Kvaesito use Frecency to rank the results. This is the same algorithm used in the Firefox web browser to calculate how to suggest results from the user's browser history.

If Kvaesito were to suggest things based on the user's local search history, I think it would significantly improve search suggestions and reduce the amount of time it takes to find something.

Related: #1609