Segmentation crash caused by setting unavailable_server_retry_responses in parent.config

[jasmine-nahrain]

Got this error when adding unavailable_server_retry_responses to the parent config

(lldb) bt
* thread #2, name = '[ET_NET 0]', stop reason = EXC_BAD_ACCESS (code=1, address=0x0)
  * frame #0: 0x000000010016aa78 traffic_server`std::__1::vector<int, std::__1::allocator<int> >::begin[abi:v15006](this=0x0000000000000000) at vector:1420:33
    frame #1: 0x000000010016a9e8 traffic_server`SimpleRetryResponseCodes::contains(this=0x0000000000000000, code=502) at ParentSelection.h:103:32
    frame #2: 0x000000010016a7c4 traffic_server`ParentResult::response_is_retryable(this=0x000000011a10da28, retry_type=PARENT_RETRY_SIMPLE, response_code=HTTP_STATUS_BAD_GATEWAY) const at ParentSelection.h:276:50
    frame #3: 0x0000000100169e88 traffic_server`response_is_retryable(s=0x000000011a10d108, response_code=HTTP_STATUS_BAD_GATEWAY) at HttpTransact.cc:398:25
    frame #4: 0x000000010015581c traffic_server`simple_or_unavailable_server_retry(s=0x000000011a10d108) at HttpTransact.cc:417:11
    frame #5: 0x0000000100152544 traffic_server`HttpTransact::handle_response_from_parent(s=0x000000011a10d108) at HttpTransact.cc:3627:3
    frame #6: 0x000000010013de3c traffic_server`HttpTransact::HandleResponse(s=0x000000011a10d108) at HttpTransact.cc:3491:5
    frame #7: 0x00000001000dea30 traffic_server`HttpSM::call_transact_and_set_next_state(this=0x000000011a10d000, f=0x0000000000000000)(HttpTransact::State*)) at HttpSM.cc:7630:5
    frame #8: 0x00000001000e8568 traffic_server`HttpSM::handle_api_return(this=0x000000011a10d000) at HttpSM.cc:1713:5
    frame #9: 0x00000001000da128 traffic_server`HttpSM::do_api_callout(this=0x000000011a10d000) at HttpSM.cc:423:5
    frame #10: 0x00000001000efa0c traffic_server`HttpSM::state_read_server_response_header(this=0x000000011a10d000, event=100, data=0x0000000128887938) at HttpSM.cc:2150:7
    frame #11: 0x00000001000d99f4 traffic_server`HttpSM::main_handler(this=0x000000011a10d000, event=100, data=0x0000000128887938) at HttpSM.cc:2757:5
    frame #12: 0x0000000100005594 traffic_server`Continuation::handleEvent(this=0x000000011a10d000, event=100, data=0x0000000128887938) at I_Continuation.h:227:12
    frame #13: 0x00000001004a8f90 traffic_server`read_signal_and_update(event=100, vc=0x0000000128887740) at UnixNetVConnection.cc:83:24
    frame #14: 0x00000001004a8468 traffic_server`read_from_net(nh=0x000000011816c400, vc=0x0000000128887740, thread=0x0000000118168000) at UnixNetVConnection.cc:323:11
    frame #15: 0x00000001004a77ec traffic_server`UnixNetVConnection::net_read_io(this=0x0000000128887740, nh=0x000000011816c400, lthread=0x0000000118168000) at UnixNetVConnection.cc:852:3
    frame #16: 0x0000000100494e7c traffic_server`NetHandler::process_ready_list(this=0x000000011816c400) at UnixNet.cc:421:11
    frame #17: 0x0000000100495648 traffic_server`NetHandler::waitForActivity(this=0x000000011816c400, timeout=60000000) at UnixNet.cc:558:3
    frame #18: 0x00000001004dc2c0 traffic_server`EThread::execute_regular(this=0x0000000118168000) at UnixEThread.cc:292:14
    frame #19: 0x00000001004dc7a8 traffic_server`EThread::execute(this=0x0000000118168000) at UnixEThread.cc:353:11
    frame #20: 0x00000001004da654 traffic_server`spawn_thread_internal(a=0x00006000020cf840) at Thread.cc:78:12
    frame #21: 0x0000000185eb1034 libsystem_pthread.dylib`_pthread_start + 136

To reproduce:

1. Set unavailable_server_retry_responses in parent.config
   1. e.g. unavailable_server_retry_responses=“502”
2. parent_retry can be any of the options