[developers] parse chart display

[Ben Waldron]

I have made a small change to the parse chart display in the LKB (in the 
case when *characterize-p* is set).

The span of each token in the chart is now displayed along with the text 
of the token. This is in order to make the structure of the token 
lattice visible.  E.g.

 0-1 thins      
 0-1 this                     <<== edge generated via spelling 
correction regex
 1-2 is
 2-3 good

when parsing "thins is good".

Note also that the token/spelling chart (tchart) can be examined with 
the command print-tchart().

- Ben