[BUG] Agents can be registered with Text Embedding models

[rithin-pullela-aws]

What is the bug?
This is an extension of #4540
The Agent Registration goes through when we provide a Text Embedding model as a Model for Agent Registration.
The Agent created this way fails during Agent Execution with a non readable error:

{
    "status": 500,
    "error": {
        "type": "ClassCastException",
        "reason": "System Error",
        "details": "class org.opensearch.ml.common.dataset.remote.RemoteInferenceInputDataSet cannot be cast to class org.opensearch.ml.common.dataset.TextDocsInputDataSet (org.opensearch.ml.common.dataset.remote.RemoteInferenceInputDataSet and org.opensearch.ml.common.dataset.TextDocsInputDataSet are in unnamed module of loader java.net.FactoryURLClassLoader @1f0e2bdc)"
    }
}

How can one reproduce the bug?
Steps to reproduce the behavior:

1. Create a Local Text Embedding Model:

POST /_plugins/_ml/models/_register
{
  "name": "huggingface/sentence-transformers/paraphrase-mpnet-base-v2",
  "version": "1.0.1",
  "model_format": "TORCH_SCRIPT"
}

2. Deploy the model:

POST /_plugins/_ml/models/<Model ID>/_deploy

3. Create an Agent with this model:

POST /_plugins/_ml/agents/_register
{
    "name": "Agent with MCP tools",
    "type": "conversational",
    "description": "Some Agent",
    "llm": {
        "model_id": "<Model ID from Step 1 and 2>",
        "parameters": {
            "max_iteration": 15
        }
    },
    "memory": {
        "type": "conversation_index"
    },
    "parameters": {
        "_llm_interface": "openai/v1/chat/completions"
    },
    "tools": [
        {
            "type": "ListIndexTool"
        },
        {
            "type": "IndexMappingTool"
        }
    ],
    "app_type": "os_chat"
}

Observe that this step succeeds and the model is registered

4. Execute the Agent:

POST /_plugins/_ml/agents/<Agent ID>/_execute
{
    "parameters": {
        "system_prompt": "helper",
        "question": "hello",
        "verbose": true
    }
}

The response would look like this:

{
    "status": 500,
    "error": {
        "type": "ClassCastException",
        "reason": "System Error",
        "details": "class org.opensearch.ml.common.dataset.remote.RemoteInferenceInputDataSet cannot be cast to class org.opensearch.ml.common.dataset.TextDocsInputDataSet (org.opensearch.ml.common.dataset.remote.RemoteInferenceInputDataSet and org.opensearch.ml.common.dataset.TextDocsInputDataSet are in unnamed module of loader java.net.FactoryURLClassLoader @1f0e2bdc)"
    }
}

What is the expected behavior?
Option 1: Improve the validation during Agent Registration, validate the models and tools.

Option 2: Have a better, more readable Error message during Agent execution.

What is your host/environment?

• OS: [e.g. iOS]
• Version [e.g. 22]
• Plugins

Do you have any screenshots?
If applicable, add screenshots to help explain your problem.

Do you have any additional context?
Add any other context about the problem.