MD4 checksum fix

[cbarratt at users.sourceforge.net]

> I've not heard from you on the adaptive checksum length patch.
> I shall be committing it shortly subject to objections or
> further discussion.

Sorry, I've been extremely busy (http://biz.yahoo.com/bw/030326/265186_1.html).
I've been following the discussion and the patch looks good.

> I would like to see the MD4 checksums fixed as well.  We are
> very close to the upper limit on protocol versions for
> deployed versions of rsync.  Therefore, i would like to
> minimize protocol increments for a while.  In any other
> circumstance i wouldn't suggest doing so but i think it
> would be a good idea to integrate these two fixes in one
> protocol bump.

I agree, they should be done together.  I don't have my original
patch but I can reimplement it with the correct remote_version
dependence and send it in the next couple of days (by Thursday
evening).  My intent is the minimal set of changes, rather than
changing the internals.

Craig