"dev" preset fails with "port already in use" for etcd

I have concierge 1.8.0 on 26.04 multipass vm,
```
Name Version Rev Tracking Publisher Notes
concierge 1.8.0 110 latest/stable jnsgruk✪ classic
```

Installing the `dev` preset fails at the `k8s bootstrap` stage for port already in use for etcd,
```
ubuntu@cos:~$ sudo concierge prepare -p dev
time=2026-09-03T14:19:43.694+03:00 level=INFO msg="Preset selected" preset=dev
time=2026-09-03T14:19:44.858+03:00 level=INFO msg="Refreshed snap" snap=charmcraft
time=2026-09-03T14:19:45.334+03:00 level=INFO msg="Installed apt package" package=gnome-keyring
time=2026-09-03T14:19:45.599+03:00 level=INFO msg="Refreshed snap" snap=jq
time=2026-09-03T14:19:45.602+03:00 level=INFO msg="Installed apt package" package=python3-pip
time=2026-09-03T14:19:45.879+03:00 level=INFO msg="Installed apt package" package=python3-venv
time=2026-09-03T14:19:46.205+03:00 level=INFO msg="Refreshed snap" snap=yq
time=2026-09-03T14:19:46.846+03:00 level=INFO msg="Refreshed snap" snap=rockcraft
time=2026-09-03T14:19:47.556+03:00 level=INFO msg="Refreshed snap" snap=snapcraft
time=2026-09-03T14:19:48.353+03:00 level=INFO msg="Refreshed snap" snap=jhack
time=2026-09-03T14:19:49.498+03:00 level=INFO msg="Refreshed snap" snap=astral-uv
time=2026-09-03T14:19:50.335+03:00 level=INFO msg="Refreshed snap" snap=k8s
time=2026-09-03T14:19:51.047+03:00 level=INFO msg="Refreshed snap" snap=lxd
time=2026-09-03T14:19:51.504+03:00 level=INFO msg="Refreshed snap" snap=kubectl
time=2026-09-03T14:19:51.755+03:00 level=INFO msg="Prepared provider" provider=lxd
Command: k8s bootstrap
Output:
Bootstrap config verification failed: pre-init checks failed for node: Encountered error(s) while verifying port availability for Kubernetes services: port 2379 (needed by: etcd) is already in use
port 2380 (needed by: etcd-peer) is already in useCommand: k8s bootstrap
Output:
Bootstrap config verification failed: pre-init checks failed for node: Encountered error(s) while verifying port availability for Kubernetes services: port 2379 (needed by: etcd) is already in use
port 2380 (needed by: etcd-peer) is already in useCommand: k8s bootstrap
Output:
Bootstrap config verification failed: pre-init checks failed for node: Encountered error(s) while verifying port availability for Kubernetes services: port 2379 (needed by: etcd) is already in use
port 2380 (needed by: etcd-peer) is already in useCommand: k8s bootstrap
Output:
Bootstrap config verification failed: pre-init checks failed for node: Encountered error(s) while verifying port availability for Kubernetes services: port 2379 (needed by: etcd) is already in use
port 2380 (needed by: etcd-peer) is already in useCommand: k8s bootstrap
Output:
Bootstrap config verification failed: pre-init checks failed for node: Encountered error(s) while verifying port availability for Kubernetes services: port 2379 (needed by: etcd) is already in use
port 2380 (needed by: etcd-peer) is already in useCommand: k8s bootstrap
```
and so it goes.

The error says it is needed by etcd, but it is etcd that is already using it:
```
ubuntu@cos:~$ sudo lsof -nP -i | grep -E ":2379|:2380"
etcd 1437 root 3u IPv4 16335 0t0 TCP 127.0.0.1:2380 (LISTEN)
etcd 1437 root 6u IPv4 16339 0t0 TCP 127.0.0.1:2379 (LISTEN)
etcd 1437 root 11u IPv4 17055 0t0 TCP 127.0.0.1:45224->127.0.0.1:2379 (ESTABLISHED)
etcd 1437 root 12u IPv4 16344 0t0 TCP 127.0.0.1:2379->127.0.0.1:45224 (ESTABLISHED)
```

Reproduce the failure with `sudo concierge prepare -p dev` and follow the `k8s bootstrap` entry point, using the reported etcd listeners on ports 2379 and 2380 as the starting evidence. Trace how the dev preset handles the existing etcd process and verify the command can complete Kubernetes bootstrap without the false port conflict.

Written by the indexing model from the issue text.