Improve stable-rt test coverage and notification reliability

## Context

Investigation of the stable-rt results following user feedback identified several infrastructure and notification gaps.

Observed issues include:
• Notification subscriptions for stable-rt require updating.
• Maintaining branch subscriptions manually can allow new branches to be missed.
• Some rt-tests jobs appear to fail with “Unsupported url protocol scheme”, potentially related to rt-tests with NFS boot.
• A trixie-rt armhf rootfs is missing, preventing or limiting rt-tests coverage on boards such as imx6q-sabrelite.

These gaps mean that the available rt-tests results may represent only a subset of the hardware where we expect the tests to run.

## Goal

Improve the reliability and coverage of stable-rt testing and ensure that relevant results are correctly included in the notification pipeline.

Tasks / points to investigate
• Update the current stable-rt notification subscriptions.
• Investigate automatically keeping branch subscriptions up to date.
• Investigate “Unsupported url protocol scheme” failures involving rt-tests and NFS boot.
• Add or restore the required trixie-rt armhf rootfs.
• Identify where rt-tests are expected to run versus where they are currently producing usable results.

No files, tests, or entry points are named. Start by tracing the stable-rt notification subscriptions and rt-tests job configuration, then investigate the NFS boot URL error and the missing trixie-rt armhf rootfs. Done means subscriptions stay current, relevant hardware produces usable results, and stable-rt notifications include them reliably.

Written by the indexing model from the issue text.