[Reporting] Add an error code field to explain the nature of the job failure

[tsullivan]

Part 1: A report job can fail for a number of reasons. The code tries to capture any error that is available, but the nature of some errors makes that infeasible:

1. User shut down Kibana while a report job was executing.
   • We should try to utilize the stop cycle of Kibana plugins: if there is a job running, then fail the job with a specific error message
   • This may still not be feasible: it's not clear if the job can be updated in ES while Kibana is shutting down
2. The system ran out of memory / resources, or a worker thread ran out of memory
   • This is only feasible if Kibana is still running after the out-of-memory error
3. If kibana.yml has xpack.screenshotting.networkPolicy.rules defined, Reporting shuts down the browser when a URL that violates the policy is encountered

Part 2: Today, a security error is ignored if encountered during CSV report execution. This creates confusion if authentication worked at the beginning of execution, but stopped working during execution - which can happen when using authentication tokens that have a short expiration. We should handle this event specifically with an error code. When this happens, the user should be able to download as much CSV as could be compiled, so the report job status should be completed_with_warnings.

In total, this issue covers adding codes for the following events:

• Kibana is intentionally stopped during report execution [Reporting] Capture Kibana stopped error #127017
• Chromium being unable to launch (whether its because of low memory or lacking OS dependencies)
• Worker thread ran out of memory, when that option is available [Reporting] Give users better information when pdf fails #126171
• Authentication expired during report execution [Reporting] Capture CSV error codes #125812

The error code will be saved as a new field in the report job mapping. It will be a mapped type, allowing us to aggregate by different types of errors. That becomes useful for telemetry and Monitoring of Reporting.

[elasticmachine]

Pinging @elastic/kibana-app-services (Team:AppServicesUx)

[jloleysens]

I think the only remaining work is to add these error_codes to our telemetry payloads, then wait until it hits production. Opened an issue for doing so: #127460