[WIP] Support additional output formats for sparse models

[yuye-aws]

Description

This PR supports the sparse model return token_id instead of the token itself. Here is the steps to test this PR:

First, register a sparse model.

POST /_plugins/_ml/models/_register?deploy=true
{
  "name": "amazon/neural-sparse/opensearch-neural-sparse-encoding-doc-v2-distill",
  "version": "1.0.0",
  "model_format": "TORCH_SCRIPT"
}

After deploying it, you can test with the following prediction.

POST _plugins/_ml/models/c6T08ZYBhR-bMOVBD8AJ/_predict
{
  "text_docs": ["hello world"],
  "parameters": {
    "sparse_encoding_format": "int"
  }
}

You'll obtain the following results:

{
  "inference_results": [
    {
      "output": [
        {
          "name": "output",
          "dataAsMap": {
            "response": [
              {
                "999": 0.07596276,
                "1007": 0.23728506,
                "1010": 0.2664351,
                "1012": 0.51118666,
                "1027": 0.07092948,
                "1028": 0.09307969,
                "1029": 0.7082903,
                "1030": 0.14336383,
                "1033": 0.21647291,
                ...
              }
            ]
          }
        }
      ]
    }
  ]
}

Related Issues

Resolves #[Issue number to be closed when this PR is merged]

Check List

• New functionality includes testing.
• New functionality has been documented.
• API changes companion pull request created.
• Commits are signed per the DCO using --signoff.
• Public documentation issue/PR created.

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.