clarify token frequency vs. relative frequency

[dwhieb]

In the Statistics tab in the search results, the frequency of each wordform is displayed as relative frequency (per million words), with the raw token frequency in parentheses immediately following:

There's no indication in the interface of what these numbers indicate. It would be good to separate them into two distinct, labeled columns if possible. Otherwise you could adjust the column header to be clearer.