kube-vip itself does not really care what IP address you use, as long as it is 'free' and in the same subnet your chosen interface is in.
Suppose eth0 on the master nodes looks like this:
// master node 1
eth0 inet addr:192.168.1.11
bcast:192.168.1.255
mask:255.255.255.0
// master node 2
eth0 inet addr:192.168.1.12
bcast:192.168.1.255
mask:255.255.255.0
// master node 3
eth0 inet addr:192.168.1.13
bcast:192.168.1.255
mask:255.255.255.0
You can now use 192.168.1.X as the VIP. 192.168.1.67 would work, and so would 192.168.1.182 or 192.168.1.10 (assuming the addresses are 'free'). None of these IP addresses is 'more valid' than the others.
I am not sure how all of this works on digitalocean (since I have never used it), but you will probably have to assign an additional IP address that all nodes can use. On linode (similar platform) for example, there is an option called 'IP sharing', which allows you to share one IP between multiple nodes.
But all that aside, you probably do not even need kube-vip, as digitalocean offers managed TCP load balancers which is probably what you are looking for. I have even found a tutorial on how to set up an HA control plane with digitalocean's managed load balancer.
