Recreates a previously-run workflow execution that will only start executing from the last known failure point.
In Recover mode, users cannot change any input parameters or update the version of the execution.
This is extremely useful to recover from system errors and byzantine faults like - Loss of K8s cluster, bugs in platform or instability, machine failures,
downstream system failures (downstream services), or simply to recover executions that failed because of retry exhaustion and should complete if tried again.
See admin.ExecutionRecoverRequest for more details.
/api/v1/executions/recover