FAT has a somewhat obnoxious rule: byte E5 in the first position is
encoded as 05.  If it was *always* that case we could simply fold it
into the codepage tables, but that rule applies only to the first
character.

Note that in the common codepages (incl. 437), E5 is lower-case sigma
(σ), which we currently upper case to upper case sigma (Σ, E4)
preventing a match -- which DOS *doesn't* do (it thinks of the Greek
characters as symbols, not subject to case distinctions.)  Putting
that special case in the codepage-generation scripts seems too ugly,
though.

We have made it this far without it, but it seems like a disaster
waiting to happen.  We can afford the extra four bytes in sector 1.

Mechanically rename the writestr functions:

	writestr   -> writestr_early
	cwritestr  -> writestr

... to reflect which function is more appropriate to call in normal
circumstances.

Print the interrupt number, AX, and CS:IP of the offending system
call.

search_dos_dir returns attributes in DL, so we shouldn't push/pop DX
around this function!

Make the codepage data all derive from the Unicode tables, and create
files that could be dynamically loaded in the future.

Initial support for VFAT long filenames; currently hard-coded to
codepage 865, no support for other codepages or for accessing files
with names which aren't in the current codepage.  This hopefully
shouldn't be an issue for SYSLINUX applications.

Move source files out of the root directory; the root is a mess and
has become virtually unmaintainable.  The Syslinux core now lives in
core/; the Linux and generic utilities has moved into utils/, and
copybs.com has moved into dos/; it had to go somewhere, and it seemed
as good a place as any.