RumbleML IndexToString does not work

Fix the problem with indexToString. The following example adapted from the docs does not work properly

let $raw-data := parallelize((
    {"id": 0, "category": "a", "categoryIndex": 1},
    {"id": 1, "category": "b", "categoryIndex": 2},
    {"id": 2, "category": "c", "categoryIndex": 3},
    {"id": 3, "category": "a", "categoryIndex": 1},
    {"id": 4, "category": "a", "categoryIndex": 1},
    {"id": 5, "category": "c", "categoryIndex": 3}
))

let $data := annotate(
    $raw-data
    { "id": "integer", "category": "string", "categoryIndex": "integer" }
)

let $transformer := get-transformer("IndexToString")
for $result in $transformer(
    $data,
    { "inputCol": "categoryIndex", "outputCol": "originalCategory"}
)
return {
    "label": $result.label,
    "category": $result.category,
    "categoryIndex": $result.categoryIndex,
    "originalCategory": $result.originalCategory
}

An error has occurred: org.apache.spark.ml.attribute.UnresolvedAttribute$ cannot be cast to org.apache.spark.ml.attribute.NominalAttribute
We should investigate this 🙈. Please contact us or file an issue on GitHub with your query.

Start by running the supplied RumbleML query and comparing its IndexToString call with the linked Spark ML example. Trace the get-transformer("IndexToString") entry point and the metadata used for categoryIndex; done means the query completes and originalCategory contains the expected category values.