This is an unsupported configuration. While yes theoretically you can install the product on each cluster node. But the product is not Cluster Aware we're unable to perform log truncation because access to both nodes is needed. What would happen if the roles were moved to another node in the cluster? Also, restoring to a Cluster will not work either, however just some internal skunk work testing that we've done is you 'can' restore to a standalone SQL server. I should however reiterate this is unsupported.
As we announced at VeeamON last week, VBR v10 and VAW3.0 will add native agent management as well as the support for Windows Failover Clusters ie - SQL AOG.