xen/intr: mark several pieces unsigned

[ehem_freebsd_m5p.com]

Summary

Fix a fair number of signed/unsigned mismatches. Since these should
always be positive, unsigned is the way to go. While such mismatches
are rampant in the FreeBSD kernel, try to reduce the number in the Xen
code.

[ehem_freebsd_m5p.com]

Trying to reduce the number is a pretty gargantuan task right now. Getting rid of signed/unsigned issues also effects D30598, D30599, D30936, and D30997. I haven't fully narrowed it down, but I suspect CPU_FOREACH() needs the variable to be signed and unsigned at the same time (compares with mp_maxid which is unsigned, but CPU_ABSENT() assumes a signed argument).

[ehem_freebsd_m5p.com]

Trying to clean these out and two more show up.

[mhorne]

I dislike this change, because truly it fixes nothing. It is just churn to appease a sense of code correctness, and code churn is not free.

[ehem_freebsd_m5p.com]

Finish full checking. The prototype for ffs*() seems *really* bad, it needs to match POSIX, but that is still impressively bad.