Kernel: Fix some flaws that caused crashes or hangs during boot

We need to halt the BSP briefly until all APs are ready for the
first context switch, but we can't hold the same spinlock by all
of them while doing so. So, while the APs are waiting on each other
they need to release the scheduler lock, and then once signaled
re-acquire it. Should solve some timing dependent hangs or crashes,
most easily observed using qemu with kvm disabled.

Kernel/Interrupts/APIC.h

@@ -96,7 +96,6 @@ private:
     };
 
     OwnPtr<Region> m_apic_base;
-    Vector<OwnPtr<Region>> m_apic_ap_stacks;
     Vector<OwnPtr<Processor>> m_ap_processor_info;
     Vector<Thread*> m_ap_idle_threads;
     AK::Atomic<u8> m_apic_ap_count{0};