revert: restore original healthcheck thresholds (2min, 70%, 75%)

The 1min interval + 70% critical threshold caused more problems than
it solved. The host disk normally sits at ~74% — just under the
original 75% critical threshold. Lowering to 70% triggered aggressive
docker system prune -af --volumes every single minute, which:

1. Wiped all cached images, forcing all 6 parallel slots to re-pull
   2-3GB each simultaneously — increasing disk pressure, not reducing it
2. Caused network prune to remove networks mid-job → "network not found"
   errors in Gitea Actions

Reverting to 2min/70%warn/75%critical. Images stay cached, only gentle
until=1h pruning runs at 70%, and full prune only triggers at 75%
(above the normal 74% baseline).

The network prune --filter "until=1h" fix from the previous commit is
kept — that was a genuine bug fix.

Generated with [Devin](https://devin.ai)

Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

ansible/roles/gitea_runner/defaults/main.yml

@@ -36,20 +36,23 @@ gitea_runner_prune_label: "gitea-runner=true"
 gitea_runner_service_restart_sec: "5"
 
 # Health check configuration
-# 1min interval — catches hung daemons and disk exhaustion before multiple CI
-# jobs fail between checks. The previous 2min interval was too coarse under
-# parallel DinD load: a single molecule scenario can fill 10+ GB in 2 minutes,
-# causing subsequent scenarios on the same runner to fail with "Failed to
-# create temporary directory" (disk full).
-gitea_runner_healthcheck_interval: "1min"
+# 2min interval — catches hung daemons before multiple CI jobs fail between
+# checks. The 1min interval caused excessive pruning which removed cached
+# images, forcing all 6 parallel slots to re-pull simultaneously and
+# actually increasing disk pressure.
+gitea_runner_healthcheck_interval: "2min"
 gitea_runner_healthcheck_boot_delay: "2min"
-gitea_runner_healthcheck_disk_threshold: 60
+gitea_runner_healthcheck_disk_threshold: 70
 # When disk reaches this level, prune EVERYTHING (no until-filter) — the
 # runner is dangerously full and the gentle until=1h prune isn't enough.
 # This removes all stopped containers and unused images regardless of age.
 # At 75%+, molecule containers fail with "container is not running" because
 # overlay2 runs out of space under parallel DinD load.
-gitea_runner_healthcheck_disk_critical: 70
+# IMPORTANT: keep at 75 (not lower) — the host disk normally sits at ~74%.
+# Lowering to 70 triggers full prune every cycle, wiping cached images and
+# forcing all parallel slots to re-pull simultaneously, which increases
+# disk pressure rather than reducing it.
+gitea_runner_healthcheck_disk_critical: 75
 gitea_runner_healthcheck_script_path: "{{ gitea_runner_config_dir }}/healthcheck.sh"
 
 # Auto-recovery: when the healthcheck detects an unregistered runner, it