Improve UX of manual allocation health setting during deployment

During a rolling upgrade of a job with "health_check" set to "manual", we encounter two major difficulties :
- getting the full deployment UUID is complicated : the only way we found was to get the short UUID from the Web UI, and then to make a GET request to /v1/deployments?prefix=. Is there another way?
- manually setting the health of the allocation must be done via the HTTP API : /v1/deployment/allocation-health/:deployment_id with a json payload which is not straightforward. Would it be possible to do it via the CLI? Like nomad promote for canary updates

### Nomad version
Nomad v0.7.1 (0b295d399d00199cfab4621566babd25987ba06e)

### Operating system and Environment details
Ubuntu 16.04 - remote cluster

Start with the deployment UUID lookup using GET /v1/deployments?prefix= and the allocation-health endpoint described in the issue, then review existing deployment-related CLI commands. Done means manual allocation health can be performed through a clear CLI workflow without constructing the current HTTP request payload by hand.

Written by the indexing model from the issue text.