Add some DBC checks in dighost; fix race between clear_query and send_done in dighost
- Add magic to dig_query_t and some DBC checks
- There was a race between clear_query and send_done - if there's a packet in flight and we're waiting to get send_done launched, and in the meantime clear_query is called, we'll try to traverse an invalid list. This behaviour is really exposed with new taskmanager/socket code.
Closes #650 (closed)
Edited by Evan Hunt