Fix nvidia-targeted-stress failing with DCGM_ST_IN_USE (-34)

nvvs (DCGM validation suite) survives when dcgmi is killed mid-run,
leaving the GPU occupied. The next dcgmi diag invocation then fails
with "affected resource is in use".

Two-part fix:
- Add nvvs and dcgmi to KillTestWorkers patterns so they are cleaned
  up by the global cancel handler
- Call KillTestWorkers at the start of RunNvidiaTargetedStressValidatePack
  to clear any stale processes before dcgmi diag runs

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

audit/internal/platform/kill_workers.go

@@ -15,6 +15,10 @@ var workerPatterns = []string{
 	"stress-ng",
 	"stressapptest",
 	"memtester",
+	// DCGM diagnostic workers — nvvs is spawned by dcgmi diag and survives
+	// if dcgmi is killed mid-run, leaving the GPU occupied (DCGM_ST_IN_USE).
+	"nvvs",
+	"dcgmi",
 }
 
 // KilledProcess describes a process that was sent SIGKILL.

audit/internal/platform/sat.go

@@ -382,6 +382,13 @@ func (s *System) RunNvidiaTargetedStressValidatePack(ctx context.Context, baseDi
 	if err != nil {
 		return "", err
 	}
+	// Kill any lingering nvvs/dcgmi processes from a previous interrupted run
+	// before starting — otherwise dcgmi diag fails with DCGM_ST_IN_USE (-34).
+	if killed := KillTestWorkers(); len(killed) > 0 && logFunc != nil {
+		for _, p := range killed {
+			logFunc(fmt.Sprintf("pre-flight: killed stale worker pid=%d name=%s", p.PID, p.Name))
+		}
+	}
 	return runAcceptancePackCtx(ctx, baseDir, "gpu-nvidia-targeted-stress", []satJob{
 		{name: "01-nvidia-smi-q.log", cmd: []string{"nvidia-smi", "-q"}},
 		{