Run pkg/jobs at frozen phase in upgrade_monitor.go

**Epic:** #1080

In `internal/daemon/consensus/upgrade_monitor.go` `handleExecute()`, after the CN is confirmed stopped (`now_frozen.mf` received) and before upgrade scripts are executed, call:

```go
jobs.RunPhaseJobs(ctx, dynamicClient, k8sClient, jobsDir, operationID, namespace, jobs.Upgrade, jobs.PhaseUpgradeFrozen)
```

## Acceptance Criteria
- [ ] Jobs run after CN stopped and before upgrade scripts
- [ ] Abort failure halts upgrade and sets DaemonResult=Failed (see Story 4.5)
- [ ] Ignore failure logs warning; upgrade scripts proceed
- [ ] Covered in `upgrade_monitor_test.go`

In internal/daemon/consensus/upgrade_monitor.go, inspect handleExecute() around receipt of now_frozen.mf and the upgrade scripts. Run the upgrade_monitor_test.go tests first, then cover the job call and its abort and ignore failure behavior. Done means frozen-phase jobs run in the required order, abort failures stop the upgrade with DaemonResult=Failed, and ignore failures log a warning while scripts proceed.

Written by the indexing model from the issue text.