Improve NFR tests

For scale NFR tests, we currently only collect CPU and memory metrics for NGF pods. We should extend this to include NGINX pods as well, so we have full visibility across the system.

Some TTR images are also not being generated, this needs to be fixed.

Notes: Investigate whether other tests (e.g., longevity) need to be updated as part of this effort.