[PATCH] acl: fix the value of the trans table

[Huichao Cai]

Maybe that's not a problem, because ACL can handle idle
packets normally, so I can't reproduce this issue with
external test cases at the moment, but I'm just describing
what idle nomatch node has to do if it wants to implement
an effect pointing to itself:

The reason why RTE_ACL_NODE_SINGLE node can point to itself
(regardless of the value of the packet data) is that it relies
on the node_index high 32-bit assignment to the maximum boundary
value (RTE_ACL_QUAD_SINGLE), and the idle nomatch node has the
same properties, so it should also be like RTE_ACL_NODE_SINGLE
node_ index instead of the current boundary value of 0, the reason
why the current idle nomatch node can point to itself is because
the idle data is 0, which has nothing to do with the node pointing
to its own attributes, it is just a coincidence of the data.