x86/smp: Fix mwait_play_dead() and acpi_processor_ffh_play_dead() noreturn behavior

x86/smp: Fix mwait_play_dead() and acpi_processor_ffh_play_dead() noreturn behavior

Fix some related issues (done in a single patch to avoid introducing
intermediate bisect warnings):

  1) The SMP version of mwait_play_dead() doesn't return, but its
     !SMP counterpart does.  Make its calling behavior consistent by
     resolving the !SMP version to a BUG().  It should never be called
     anyway, this just enforces that at runtime and enables its callers
     to be marked as __noreturn.

  2) While the SMP definition of mwait_play_dead() is annotated as
     __noreturn, the declaration isn't.  Nor is it listed in
     tools/objtool/noreturns.h.  Fix that.

  3) Similar to #1, the SMP version of acpi_processor_ffh_play_dead()
     doesn't return but its !SMP counterpart does.  Make the !SMP
     version a BUG().  It should never be called.

  4) acpi_processor_ffh_play_dead() doesn't return, but is lacking any
     __noreturn annotations.  Fix that.

This fixes the following objtool warnings:

  vmlinux.o: warning: objtool: acpi_processor_ffh_play_dead+0x67: mwait_play_dead() is missing a __noreturn annotation
  vmlinux.o: warning: objtool: acpi_idle_play_dead+0x3c: acpi_processor_ffh_play_dead() is missing a __noreturn annotation

Fixes: a7dd183f0b38 ("x86/smp: Allow calling mwait_play_dead with an arbitrary hint")
Fixes: 541ddf31e300 ("ACPI/processor_idle: Add FFH state handling")
Reported-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Paul E. McKenney <paulmck@kernel.org>
Link: https://lore.kernel.org/r/e885c6fa9e96a61471b33e48c2162d28b15b14c5.1740962711.git.jpoimboe@kernel.org