Remove timer-based runtime advance from interrupt handler

The approach of setting ec->runtime = ec->quantum inside
rb_threadptr_execute_interrupts caused a SystemStackError on Windows
when a timer interrupt fired during @main.transfer inside the base
Scheduler: preemption called scheduler.yield -> kernel_sleep(0) ->
block -> @main.transfer -> recursion.

Remove the timer-based advance entirely.  Under YARV the back-edge
counter in rb_vm_check_ints provides preemption without this.  Under
YJIT the timing-based preemption tests are already guarded with
omit/skip so they don't run.

Co-authored-by: Cursor <cursoragent@cursor.com>

Author: samuel-williams-shopify

thread.c

@@ -2719,15 +2719,6 @@ rb_threadptr_execute_interrupts(rb_thread_t *th, int blocking_timing)
                             0, 0, 0, Qundef);
 
             rb_thread_schedule_limits(limits_us);
-
-            // Advance the fiber's scheduling runtime to the quantum so that
-            // the next rb_vm_check_ints call (the next safe point) triggers
-            // rb_fiber_scheduler_maybe_preempt.  This provides timer-based
-            // fiber preemption for JIT-compiled code that bypasses the
-            // back-edge counter without executing Ruby code in interrupt context.
-            if (th->ec->thread_ptr->blocking == 0 && !th->ec->preempted) {
-                th->ec->runtime = th->ec->quantum;
-            }
         }
     }
     return ret;