Updating to match with D30648 update (update to diff, though the result is the same). Reorder xen_intr_pic_enable_source()/xen_intr_pic_disable_source() to match xen_intr.c ordering.

Reverting to previous version. Upon examination, that was not an improvement.

Fixing corruption which occurred during work. Ensuring more comment blocks get preserved from portions completely removed from xen_intr.c.

Update fixing logic inversion found during look. The return of xen_arch_intr_has_handlers() is inverted from how the KASSERT() was previously.

Figured out the unpleasant feeling I was having. The cookie is the *handler* cookie. As such the Xen interrupt should really simply be forwarding it to client code. Given this, the cookie belongs to the architecture independent side and the architecture dependent side should simply forward.

Better arrangement of enum evtchn_type. Nothing outside of the interrupt interface needs this, so reducing the visibility is good. I'm still wondering about the xen_arch_intr_remove_handler() interface. Simplest approach to clearing xi_cookie might be D31188, which ensures everything is cleared.

Identifying where the need for <machine/intr_machdep.h> actually disappears.

Switching to inlines for the paper-thin compatibility layer bits. There is only the tiniest conversion between architecture-independent and architecture-dependent for these.

Keeping things synchronized. Looks like a few stray bits snuck in last time. Need those approvals so I'm not dealing with a huge pile of local commits...

Single blank line adjustment, but keeping Phabricator synchronized.