[BUG] Deployment fails if we do not set the MungeKeySsmParameter

[deeppat]

Describe the bug
If you try to deploy the eda slurm cluster with the default config file with the MungeKeySsmParameter commented out, cdk still tries to find it in your account and fails deployment when it doesn't find the key.

For eg. my config file has it commented:

  #MungeKeySsmParameter: "/slurm/munge_key"

To Reproduce
Steps to reproduce the behavior:

1. Clone Repo
2. Call ./install.sh --prompt --cdk-cmd create
3. Bootstrap succeeds, but slurm deployment fails with the following error:

slurmminimal: creating CloudFormation changeset...

 ❌  slurmminimal failed: Error [ValidationError]: Unable to fetch parameters [/slurm/munge_key] from parameter store for this account.
    at Request.extractError (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/protocol/query.js:50:29)
    at Request.callListeners (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/sequential_executor.js:106:20)
    at Request.emit (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/sequential_executor.js:78:10)
    at Request.emit (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/request.js:686:14)
    at Request.transition (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/request.js:22:10)
    at AcceptorStateMachine.runTo (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/state_machine.js:14:12)
    at /home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/state_machine.js:26:10
    at Request.<anonymous> (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/request.js:38:9)
    at Request.<anonymous> (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/request.js:688:12)
    at Request.callListeners (/home/ec2-user/.nvm/versions/node/v16.15.0/lib/node_modules/aws-sdk/lib/sequential_executor.js:116:18) {
  code: 'ValidationError',
  time: 2022-05-12T16:48:12.685Z,
  requestId: 'xxx',
  statusCode: 400,
  retryable: false,
  retryDelay: 542.6911270021684

Expected behavior
Expect the slurm app to be deployed

Screenshots
If applicable, add screenshots to help explain your problem.

Repository Version):
14fe152

Additional context
Add any other context about the problem here.

[cartalla]

Confirmed. The ssm parameter name has a default name and it needs to be created and set if it doesn't already exist.
Otherwise, cannot give IAM permissions to the parameter so the slurm controller can read/write it.

[cartalla]

I was able to deploy a cluster using the bug's branch:

21-bug-deployment-fails-if-we-do-not-set-the-mungekeyssmparameter