perf: cut redundant work out of job setup and teardown (#1218)

Implement speedups to job start and shutdown.

- Create the job container while services are still becoming healthy, and poll their health at a flat one second instead of a 2s to 32s doubling backoff
- Pull each service image once instead of twice, and fetch a warm action cache once instead of twice
- Report the job result before reclaiming its volumes, and reap volumes stranded by a runner that died mid-job

| Step | Scenario | Before | After |
| --- | --- | --- | --- |
| Complete job | Large workspace volume | 4.2s | 0.4s |
| Set up job | One service, 2s health interval | 7.08s | 3.26s |
| Set up job | Two cached actions from github.com | 1.81s | 1.34s |
| Set up job | Two cached actions from gitea.com | 2.42s | 1.95s |
| Set up job | Two actions, cold action cache | 6.62s | unchanged |
| Set up job | Minimal job, no services or actions | 0.62s | unchanged |

Assisted-by: Claude Code:Opus 5
Reviewed-on: https://gitea.com/gitea/runner/pulls/1218
Reviewed-by: bircni <bircni@icloud.com>
Co-authored-by: silverwind <me@silverwind.io>

internal/app/run/runner_idle_cleanup_test.go

@@ -311,6 +311,7 @@ func TestRunnerOnIdleRemovesOrphanNetworks(t *testing.T) {
 	}
 
 	var swept []string
+	var sweptVolumes []string
 	var sweptCutoff time.Time
 	origRemoveOrphanNetworks := removeOrphanNetworks
 	removeOrphanNetworks = func(_ context.Context, runnerUUID string, createdBefore time.Time) error {
@@ -319,15 +320,32 @@ func TestRunnerOnIdleRemovesOrphanNetworks(t *testing.T) {
 		return nil
 	}
 	t.Cleanup(func() { removeOrphanNetworks = origRemoveOrphanNetworks })
+	origRemoveOrphanJobVolumes := removeOrphanJobVolumes
+	removeOrphanJobVolumes = func(_ context.Context, runnerUUID string, createdBefore time.Time) error {
+		sweptVolumes = append(sweptVolumes, runnerUUID)
+		assert.Equal(t, now.Add(-24*time.Hour), createdBefore)
+		return nil
+	}
+	t.Cleanup(func() { removeOrphanJobVolumes = origRemoveOrphanJobVolumes })
 
 	r := &Runner{uuid: "runner-1", cfg: cfg, now: func() time.Time { return now }}
 	r.OnIdle(context.Background())
 	assert.Equal(t, []string{"runner-1"}, swept)
+	assert.Equal(t, swept, sweptVolumes)
 	// a network of a job starting during the pass is younger than this and so out of scope
 	assert.Equal(t, now.Add(-24*time.Hour), sweptCutoff)
 
 	// a host-only runner has no daemon to sweep
+	origDockerReachable := dockerReachable
+	dockerReachable = func(context.Context) bool { return false }
+	t.Cleanup(func() { dockerReachable = origDockerReachable })
 	hostOnly := &Runner{uuid: "runner-2", cfg: &config.Config{Runner: cfg.Runner}, now: func() time.Time { return now }}
 	hostOnly.OnIdle(context.Background())
 	assert.Equal(t, []string{"runner-1"}, swept)
+	assert.Equal(t, swept, sweptVolumes)
+
+	dockerReachable = func(context.Context) bool { return true }
+	now = now.Add(time.Minute)
+	hostOnly.OnIdle(context.Background())
+	assert.Equal(t, []string{"runner-1", "runner-2"}, sweptVolumes)
 }