Describe the bug
When ingesting an Elasticsearch source into DataHub, if the index mapping includes a nested property with multiple sub-properties, DataHub fails to detect the sub-properties of the nested field after ingestion. For example, with the following mapping:

{
  "example-index": {
    "mappings": {
      "properties": {
        "addresses": {
          "type": "nested",
          "properties": {
            "address": {
              "type": "keyword"
            },
            "type": {
              "type": "keyword"
            }
          }
        },
        "id": {
          "type": "keyword"
        },
        "name": {
          "type": "keyword"
        }
      }
    }
  }
}

In this case, DataHub does not detect the address and type properties within the addresses nested field.

To Reproduce
Steps to reproduce the behavior:

1. Setup the elastic mapping provided in the example

PUT /example-index
{
  "mappings": {
    "properties": {
      "addresses": {
        "type": "nested",
        "properties": {
          "address": {
            "type": "keyword"
          },
          "type": {
            "type": "keyword"
          }
        }
      },
      "id": {
        "type": "keyword"
      },
      "name": {
        "type": "keyword"
      }
    }
  }
}

2. Go to 'Elasticsearch Source Configuration' in DataHub.
3. Click on 'Ingest Elasticsearch Index'.
4. Scroll down to the section with the nested properties in the index mapping.
5. See error: Sub-properties of the nested field are not detected.
   

Expected behavior
The sub-properties of the nested field should be detected and ingested properly by DataHub, allowing them to be visible and usable within the DataHub interface.

Desktop (please complete the following information):

• OS: Ubuntu
• Browser Chrome
• Version 126.0.6478.126 (Official Build) (64-bit)

Additional context
The source of this issue resides here: https://github.com/datahub-project/datahub/blob/master/metadata-ingestion/src/datahub/ingestion/source/elastic_search.py

I tried to find a possible solution that can be used and that worked for my situation: