Add alias and data stream to list tool

[ssunno]

Description

This PR add alias and data stream to return of ListIndicesTool.

The aliases and data streams are concepts that group multiple indices. But current version of ListIndicesTool does not support them, so it returns all indices equally, exposing the backing indices of data streams directly. To apply the intention of index designers, aliases and data streams should be returned as separate groups.

Note: Indices that belong to an alias or a data stream are excluded from the general index list to avoid duplication.

When include_detail = False, aliases and data streams return only the number of indices instead of their names.

example of ListIndicesTool (include_detail = False):

Indices, aliases, and data streams information:
{
  "indices": [
    "sonested",
    ".opendistro_security",
    ".opensearch-observability",
    ".plugins-ml-config"
  ],
  "aliases": [
    {
      "alias": "logs_current",
      "index_count": 2
    }
  ],
  "data_streams": [
    {
      "name": "logs-ds",
      "timestamp_field": {
        "name": "@timestamp"
      },
      "generation": 1,
      "status": "YELLOW",
      "template": "logs-template",
      "index_count": 1
    }
  ]
}

Issues Resolved

Resolves #56

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[rithin-pullela-aws]

Thanks for raising the PR @ssunno, I believe this is a step in the right direction. Added few comments.

[rithin-pullela-aws]

If the user is not authorized to perform the request, we might get an error response. This error should be propagated to the client making request in some form. But this would return an empty list

[rithin-pullela-aws]

Same as above, should we handle it differently?

[rithin-pullela-aws]

Same as above

[rithin-pullela-aws]

Nit:
Would not block this PR, but something to perhaps think of

On a high level for list_indices, list_aliases, and list_data_streams

Would it make sense to create a class for the responses and parse the response into the classes? We can perhaps throw some sort of error if parsing fails. That way we can skip doing checks like isInstance etc?