[Ingest Manager] Fix /agent_config/delete API 500 if default config missing

[jfsiii]

Summary

Broke the changes from b4f4051 out to a separate PR & added a test

getDefaultAgentConfigId errors if there isn't a default config. The config is added by setupIngestManager which was always called during plugin#start but is no longer.

We could add the setup call to the test/suite, but instead I changed AgentConfigService.delete to use ensureDefaultAgentConfig instead of getDefaultAgentConfigId.

ensureDefaultAgentConfig adds one if it's missing. The check in delete is to make sure we don't delete the default config. We can still do that and now we add a config if it wasn't already there (which seems like A Good Thing)

Checklist

Delete any items that are not applicable to this PR.

• Unit or functional tests were updated or added to match the most common scenarios

[elasticmachine]

Pinging @elastic/ingest-management (Team:Ingest Management)

[kibanamachine]

💚 Build Succeeded

• continuous-integration/kibana-ci/pull-request
• Commit: 73f95d0

Build metrics

✅ unchanged

To update your PR or re-run it, just comment with:
@elasticmachine merge upstream

[nchaulet]

Not sure it's the perfect solution, as it's going to eventually create the default config outside of the setup (not sure of the side effect of this)

What do you think of catching 404 of the previous call?

something like

try {
  const defaultConfigId = await this.getDefaultAgentConfigId(soClient);

  if (id === defaultConfigId) {
      throw new Error('The default agent configuration cannot be deleted');	      throw new Error('The default agent configuration cannot be deleted');
    }	    }
} catch (err) {
  // implement real test here :) 
  if (err !== 404) { throw err; }
}

[jfsiii]

closed by #73212