Implement runit support for Fenris monitoring (issue #84)

Deliver end-to-end native monitoring path under runit with existing
CLI/TUI controls and truthful status, preserving systemd behavior.

Changes:
- Add init system abstraction layer (src/fenris/init_system.py) that
  detects systemd vs runit and provides unified interface for timer
  control, on-demand collection, and service state queries
- Create runit service files (units/runit/) with completion-relative
  5-minute cadence, 2-minute boot delay, bounded execution (90s),
  no catch-up, and serialized runs via flock
- Update monitor.py to use abstraction layer instead of direct systemctl
- Update status.py to use abstraction layer for service state queries
- Update all packaging scripts (deb, rpm) for init-system-aware setup
- Update Makefile to install runit service files alongside systemd units
- Add 46 tests for init system abstraction layer

Spec: ADR 0008, §8.4, §8.5, §8.6, §8.7, §8.8
Closes #84

Co-authored-by: CommandCodeBot <noreply@commandcode.ai>

units/runit/fenris-collect/log/run

@@ -0,0 +1,9 @@
+#!/bin/sh
+# fenris-collect runit log script — service output to /var/log/fenris-collect/
+#
+# Runit automatically pipes the service's stdout/stderr to this logger's stdin.
+# The logger writes to runit's log directory for diagnostics.
+#
+# Spec: §8.8 (actionable native diagnostics)
+exec chpst -u fenris:fenris \
+    svlogd -tt /var/log/fenris-collect/

units/runit/fenris-collect/run

@@ -0,0 +1,40 @@
+#!/bin/sh
+# fenris-collect runit run script — periodic NVMe collection scheduler.
+#
+# Runit service layout:
+#   /etc/sv/fenris-collect/run        — this script (scheduler)
+#   /etc/sv/fenris-collect/log/run    — logger to /var/log/fenris-collect/
+#   /var/service/fenris-collect        — symlink to enable
+#   /etc/sv/fenris-collect/down        — marker for dormant install
+#
+# Guarantees (must match systemd timer semantics):
+#   - Initial 2-minute boot delay (sleep 120 before first collect)
+#   - Completion-relative 5-minute cadence (sleep 300 after collect)
+#   - Bounded execution (timeout 90s on each collect)
+#   - No catch-up (fixed sleep interval, no Persistent=)
+#   - Serialized runs (runsv does not restart until exit)
+#   - Serialized on-demand (flock serializes fenris-collect execution)
+#
+# Spec: §8.4, §8.5, §8.6, ADR 0008
+set -eu
+
+COLLECT_TIMEOUT=90
+BOOT_DELAY=120
+CADENCE=300
+LOCK_FILE=/var/lib/fenris/fenris-collect.lock
+
+# Ensure lock directory exists
+mkdir -p "$(dirname "$LOCK_FILE")"
+
+# Initial boot delay: sleep before first collection
+sleep "$BOOT_DELAY"
+
+# Collection loop: collect, then sleep for cadence
+while true; do
+    flock --nonblock "$LOCK_FILE" \
+        timeout "$COLLECT_TIMEOUT" nice ionice -c3 \
+        /usr/libexec/fenris/fenris-collect \
+        2>&1 || true
+
+    sleep "$CADENCE"
+done