[BUG] NeuralQueryEnricher does not work with nested queries

[IanMenendez]

NeuralQueryEnricher search processor is not working with NestedQueries
Other people are having the same issue: opensearch-project/OpenSearch#13961

how to reproduce:

PUT /_search/pipeline/test_pipeline 
{
  "request_processors": [
    {
      "neural_query_enricher" : {
        "default_model_id": "u5j0qYoBMtvQlfhaxOsa",
        "neural_field_default_id": {
           "embedding_field.knn": "uZj0qYoBMtvQlfhaYeud"
        }
      }
    }
  ]
}

GET testindex/_search?search_pipeline=test_pipeline
{
  "query": {
    "nested": {
      "score_mode": "max",
      "path": "passage_chunk_embedding",
      "query": {
        "neural": {
          "embedding_field.knn": {
            "query_text": "document"
          }
        }
      }
    }
  }
}

This should work but instead throws:

{
  "error": {
    "root_cause": [
      {
        "type": "null_pointer_exception",
        "reason": "modelId is marked non-null but is null"
      }
    ],
    "type": "null_pointer_exception",
    "reason": "modelId is marked non-null but is null"
  },
  "status": 500
}

[mingshl]

this issue should go to neural search repo. neural_query_enricher processor is in neural search repo.

[IanMenendez]

opened new issue in neural search repo as I was unable to transfer it. opensearch-project/neural-search#813