Hi,
I've not seen anything like that documented.
One thing to consider is, what precisely do you mean by "public IP range for compute instances"? A VM can use a public IP via default outbound access or NAT Gateway or Azure Firewall or load balancer or direct association or (something I'm forgetting). And if the IP is associated with VM's interface it could be MS public IP or customer public IP (aka bring-your-own-IP).
Further, VM instances are used by a ton of different Azure services under the covers. Which of these should be included/excluded? One example, there are different services whereby you set the parameters (size, quantity, scaling, etc.) for the VMs that make up a cluster. Should the public IPs that these VMs use be included/excluded?
You may want to instead consider only whitelisting specific IPs based on requests by your customers.
Please click Accept Answer and upvote if the above was helpful.
Thanks.
-TP