fix(ci): e2e nested bootstrap cluster add d8 v ssh timeout

[universal-itengineer]

Description

Add SSH keepalive options to d8 v ssh calls used in nested-cluster CI and the e2e d8 helper.
This keeps long-lived SSH sessions more stable when the VM connection is proxied through virtualization-api port-forwarding.
The following SSH options were added:

• ServerAliveInterval=15 to send periodic keepalive probes during long-running sessions.
• ServerAliveCountMax=8 to tolerate short interruptions before the SSH client gives up.
• ConnectTimeout=10 to fail stalled connection attempts quickly and avoid hanging retries.

Why do we need it, and what problem does it solve?

Nested cluster bootstrap and kubeconfig generation use d8 v ssh over a websocket-backed port-forward path.
When that path is interrupted, CI may fail with errors like websocket: close 1006, Broken pipe, or exit status 255.
Adding ServerAliveInterval, ServerAliveCountMax, and a fixed ConnectTimeout makes these SSH sessions more resilient to transient transport issues and short interruptions.

What is the expected result?

d8 v ssh sessions in CI and e2e helpers should better survive short-lived connection drops.
This should reduce flaky failures during long-running nested-cluster operations without changing the functional behavior of the tests.

Checklist

• The code is covered by unit tests.
• e2e tests passed.
• Documentation updated according to the changes.
• Changes were tested in the Kubernetes cluster manually.

Changelog entries

section: ci
type: fix
summary: add SSH keepalive options to d8-based nested cluster access in CI and e2e helpers
impact_level: low