rf_zmq_imp: revert 28399dd876 and fix unit

although the manual test with Amarisoft eNB worked fine it seems
the delay is still needed in the default case. Over 50% of the
tests failed in the nightly with:

[zmq] Error: tx time is 0.067 ms in the past (138240 < 139776)
[zmq] Error: tx time is 1.100 ms in the past (184320 < 209664)

While this usleep() should increase the pass likelihood it
still doesn't guarantee error-free runs, so we might need
to revisit it again as some stage.

lib/src/phy/rf/rf_zmq_imp.c

@@ -668,6 +668,9 @@ int rf_zmq_recv_with_time_multi(void* h, void** data, uint32_t nsamples, bool bl
     rf_zmq_info(handler->id, " - next rx time: %d + %.3f\n", ts_rx.full_secs, ts_rx.frac_secs);
     rf_zmq_info(handler->id, " - next tx time: %d + %.3f\n", ts_tx.full_secs, ts_tx.frac_secs);
 
+    // Leave time for the Tx to transmit
+    usleep((1000000UL * nsamples_baserate) / handler->base_srate);
+
     // check for tx gap if we're also transmitting on this radio
     for (int i = 0; i < handler->nof_channels; i++) {
       if (handler->transmitter[i].running) {