Imagery conversion beginnings

[Stan Shebs]

I've just committed an assortment of random changes intended to
support the mass conversion of the old textual image format
into file images.  The first change is to ximfapp, to give it
"grid" and "foreground color" options.  Together with the
background color, you can use these to collect a bunch of images,
lay them out on the display with 2-pixel wide lines separating,
then take a screenshot.

The next step is to delete the text image data in the imf file and
substitute references to the file image.  This requires offsets,
and to make the computer do the work of calculating offsets, I
added a "std" modifier that lets you specify relative positions
in the grid.  So (file "foo.gif" std 4 5) means 4th column, 5th
row, which might calculate out to a pixel position of 76,92
for a grid of 16x16 images.

Then to test it all, I converted flags.imf, and made both 8x8
and 16x16 file images.  Most of the 16x16 images are still
just magnified 8x8 images; I've improved a few of them, and
expect to have the rest done in a few days (unless somebody
else is keen for the labor).  Amusingly, the combination of
file image and modified flags.imf nets out to about 13K, just
1/3 of the size of the previous flags.imf - of course, such
small files don't have much effect on Xconq startup time
either way!

Stan