Evaluation: Sanitize SQL error messages

[vprashrex]

Is your feature request related to a problem?
The /api/v1/evaluations?limit=-1&offset=0 endpoint returns a raw SQL error to the client. This is problematic as clients should receive a proper sanitized error message instead of technical details.

Describe the solution you'd like

• Implement error handling to return a sanitized error message for SQL issues.
• Use FastAPI query parameters that enforce limits between 1 and 100, with a default of 50:

  limit: int = Query(
      default=50, ge=1, le=100, description="Maximum number of datasets to return"
  )

Original issue

Describe the bug
A clear and concise description of what the bug is.

/api/v1/evaluations?limit=-1&offset=0
this endpoint returns raw sql error, the client should not be getting errors like this, instead client should get proper sanitized error

{
    "success": false,
    "data": null,
    "error": "(psycopg.errors.InvalidRowCountInLimitClause) LIMIT must not be negative\n[SQL: SELECT evaluation_run.id, evaluation_run.run_name, evaluation_run.dataset_name, evaluation_run.type, evaluation_run.language_id, evaluation_run.providers, evaluation_run.config_id, evaluation_run.config_version, evaluation_run.dataset_id, evaluation_run.batch_job_id, evaluation_run.embedding_batch_job_id, evaluation_run.status, evaluation_run.object_store_url, evaluation_run.score_trace_url, evaluation_run.total_items, evaluation_run.score, evaluation_run.error_message, evaluation_run.organization_id, evaluation_run.project_id, evaluation_run.inserted_at, evaluation_run.updated_at \nFROM evaluation_run \nWHERE evaluation_run.organization_id = %(organization_id_1)s::INTEGER AND evaluation_run.project_id = %(project_id_1)s::INTEGER AND evaluation_run.type = %(type_1)s::VARCHAR ORDER BY evaluation_run.inserted_at DESC \n LIMIT %(param_1)s::INTEGER OFFSET %(param_2)s::INTEGER]\n[parameters: {'organization_id_1': 1, 'project_id_1': 1, 'type_1': 'text', 'param_1': -1, 'param_2': 0}]\n(Background on this error at: https://sqlalche.me/e/20/9h9h)",
    "errors": null,
    "metadata": null
}

use fastapi query for this which will handle all this

    limit: int = Query(
        default=50, ge=1, le=100, description="Maximum number of datasets to return"
    ),

To Reproduce
Steps to reproduce the behavior:

1. Go to '...'
2. Click on '....'
3. Scroll down to '....'
4. See error

Expected behavior
A clear and concise description of what you expected to happen.

Screenshots
If applicable, add screenshots to help explain your problem.

Additional context
Add any other context about the problem here.