Given that hosters of a VPS often gives a big /48, /56 or /64 ipv6 subnet to a VPS I do wonder if the BridgeLine for ipv6 could benefit from that?
ip6tables -t nat -I PREROUTING -p tcp -j DNAT --to-destination [obfs4 address]
/usr/sbin/ip6tables-save > /etc/iptables/rules.v6
all incoming TCPv6 packets would be flow to the bridge address. This idea is implemented in . BTW there I do change the factory default ipv6 address into a random one too.
 tor-relays/network.yaml at main · toralf/tor-relays · GitHub