Currently development code can break if a branch with backwards incompatible changes is merged to main, because avo cli always pulls the latest main/branch code.

By introducing incremental schema versions we could stick to a certain Avo version in the clients and allow different development teams to catch up with analytics in their own pace.