sleep compromise

[Yuhong Bao]

Remember about the sleep veto debate?
I think we should compromise by:
* Notifing when a program vetos or takes too long to respond to the
PBT_APMSUSPEND or PBT_APMQUERYSUSPEND omessage.
* Giving the user choice about whether to allow vetoing and allow the user
to set the limit on how long programs can take to respond to the
PBT_APMSUSPEND or PBT_APMQUERYSUSPEND message before the user is notified if
vetoing is allowed or if not, force the computer to sleep.
* Allowing programs to detect whether the lid is closed.