[ML] Add support for per-partition categorization jobs

[peteharverson]

Meta issue outlining the steps to add support for per-partition categorization anomaly detection jobs in the ML UI.

• Add a control to the categorization job wizard to enable per_partition_categorization for the job (switch should default to false)
• If per_partition_categorization is enabled, display an input field allowing the user to set the partitioning field for the job.
• If per_partition_categorization is enabled, display an extra checkbox/switch allowing the user to set stop_on_warn (default to false)
• Add a new Kibana endpoint for retrieving categorizer_stats documents (will search for job_id: whatever and result_type: categorizer_stats)
• Add / edit categorization Kibana endpoints to support per partition categorization to add partition_field_value to search terms when searching for categorization results from jobs with per-partition categorization enabled
• Add a new Kibana endpoint so that solutions can find out which partitions we stopped categorizing (searching for categorizer_stats documents in the job's results index)
• If viewing the results of a per-partition categorization job in the Anomaly Explorer, and if stop_on_warn was enabled, add an indicator somewhere in the view if the categorization status goes to warn for a partition. This is needed to explain to the user that there may be fewer results than there could have been.
• (Nice to have) Add an indicator in the categorization job wizard if the categorization status goes to warn for a partition

[elasticmachine]

Pinging @elastic/ml-ui (:ml)