Use _sched_block_dequeue() helper in mutex_block_atomic()

Previously, mutex_block_atomic() only marked the running task as
TASK_BLOCKED, which was sufficient when scheduling selected tasks
by scanning the global task list.

Since the new scheduler is designed to select only runnable tasks
from ready_queue[], mutex blocking now also requires removing the
task’s rq_node from the corresponding ready queue, preventing the
scheduler from selecting a blocked (non-runnable/dequeued) task again.

Author: vicLin8712

kernel/mutex.c

@@ -76,9 +76,12 @@ static void mutex_block_atomic(list_t *waiters)
     /* Add to waiters list */
     if (unlikely(!list_pushback(waiters, self)))
         panic(ERR_SEM_OPERATION);
-
+    
     /* Block and yield atomically */
     self->state = TASK_BLOCKED;
+
+    /* Explicit remove list node from the ready queue */
+    _sched_block_dequeue(self);
     _yield(); /* This releases NOSCHED when we context switch */
 }