Add support for startup script per nodeset

[mr0re1]

testing done

  - id: s1
    source: modules/scripts/startup-script
    settings:
      runners:
      - type: shell
        destination: s1
        content: |
          #!/bin/bash
          echo "p1" > /tmp/secret

  - id: ns1
    source: ./community/modules/compute/schedmd-slurm-gcp-v6-nodeset
    use: [network, s1]
    settings:
      node_count_static: 1
      machine_type: n2-standard-2
      enable_placement: false # the default is: true
  
  - id: p1
    source: ./community/modules/compute/schedmd-slurm-gcp-v6-partition
    use: [ns1]
    settings:
      partition_name: p1
      exclusive: false # allows nodes to stay up after jobs are done
      
// Same for s2, ns2, p2

$ srun -N 1 -p p1 cat /tmp/secret
p1
$ srun -N 1 -p p2 cat /tmp/secret
p2