dr_replication: fix a race in the DR replication scenario

Now that we are using a new module for secrets engines verification
without retries we could run into a race:

```
╷
│ Error: exit status 1
│
│ Error: Execution Error
│
│   with module.verify_replicated_data_during_failover.enos_remote_exec.auth_login_testuser,
│   on ../../modules/verify_secrets_engines/modules/read/auth.tf line 8, in resource "enos_remote_exec" "auth_login_testuser":
│    8: resource "enos_remote_exec" "auth_login_testuser" {
│
│ failed to execute commands due to: running script:
│ [/Users/ryan/code/hashi/vault/enos/modules/verify_secrets_engines/scripts/auth-userpass-login.sh]
│ failed, due to: 1 error occurred:
│       * executing script
│ auth/userpass/login/testuser: Error making API request.
│
│ URL: PUT http://[::1]:8200/v1/auth/userpass/login/testuser
│ Code: 503. Errors:
│
│ * Vault is sealed
```

We solve this by waiting for the cluster to be fully unsealed before
attempting to login during the data read.

Signed-off-by: Ryan Cragun <[email protected]>

enos/enos-scenario-dr-replication.hcl

@@ -990,11 +990,37 @@ scenario "dr_replication" {
     }
   }
 
+  step "verify_new_primary_cluster_unsealed" {
+    description = global.description.verify_vault_unsealed
+    module      = module.vault_verify_unsealed
+    depends_on = [
+      step.wait_for_demoted_cluster_leader,
+    ]
+
+    providers = {
+      enos = local.enos_provider[matrix.distro]
+    }
+
+    verifies = [
+      quality.vault_auto_unseals_after_autopilot_upgrade,
+      quality.vault_seal_awskms,
+      quality.vault_seal_pkcs11,
+      quality.vault_seal_shamir,
+    ]
+
+    variables {
+      hosts             = step.get_secondary_cluster_ips.follower_hosts
+      vault_addr        = step.create_secondary_cluster.api_addr_localhost
+      vault_install_dir = global.vault_install_dir[matrix.artifact_type]
+    }
+  }
+
   step "verify_replicated_data_during_failover" {
     description = global.description.verify_secrets_engines_read
     module      = module.vault_verify_secrets_engines_read
     depends_on = [
-      step.wait_for_demoted_cluster_leader
+      step.wait_for_demoted_cluster_leader,
+      step.verify_new_primary_cluster_unsealed,
     ]
 
     providers = {
@@ -1025,7 +1051,9 @@ scenario "dr_replication" {
           so that secondary clusters can utilize it.
         EOF
     module      = module.generate_secondary_public_key
-    depends_on  = [step.verify_replicated_data_during_failover]
+    depends_on = [
+      step.verify_replicated_data_during_failover,
+    ]
 
     verifies = quality.vault_api_sys_replication_dr_primary_secondary_token_write