Copy fields on nested child documents? Ngram search on multiple child docs to find a specific parent - solr

Question

I am transforming JSON objects and index them in Solr 9. When resolving lists / arrays of objects, I am using nested child documents, so array elements are stored as own documents.
Now, I ran into the issue that I would like to use copy fields on nested child documents and store this value in the parent.
JSON:
{
"legalName": "Some Name",
"addresss": {
"street": "Bala Street",
"houseNr": 13,
"city": "Random City",
"postalCode": 1234,
"country": "NL"
},
"otherLegalNames": [
{
"text": "TEXT IN EN",
"lang": "EN"
},
{
"text": "TEXT IN DE",
"lang": "DE"
},
{
"text": "TEXT IN NL",
"lang": "NL"
}
]
}
When indexing this object, I flatten basic structs like address but keep arrays, e.g., otherLegalNames, and stored them through child docs.
Basically, the documents look like this (q=*:*&fl=*,[child]):
{
"id": "5493006O42CR4SHELP26",
"legalName": "Some Name",
"addresss.street": "Bala Street",
"addresss.houseNr": 13,
"addresss.city": "Random City",
"addresss.postalCode": 1234,
"addresss.country": "NL",
"otherLegalNames": [
{
"id": "5493006O42CR4SHELP26/otherLegalNames#0",
"text": "TEXT IN EN",
"lang": "EN"
},
{
"id": "5493006O42CR4SHELP26/otherLegalNames#1",
"text": "TEXT IN DE",
"lang": "DE"
},
{
"id": "5493006O42CR4SHELP26/otherLegalNames#2",
"text": "TEXT IN NL",
"lang": "NL"
}
]
}
Now I would like to search for these docs by their legalName and must therefore search in the parent legalName field but also include all text fields stored under otherLegalNames. During research, I found that copy fields are the way to go, but I am not sure how I would handle child documents with such copy fields.
My goal would be to get a searchableLegalNames field with the value: ["Some Name", "TEXT IN EN", "TEXT IN DE", "TEXT IN NL"] or similar to perform a Ngram based search on legalName including every language.
Is this possible to achieve with copy fields or are child documents not supported for this purpose? If so, how should I restructure my schema? It's really hard to flatten every legal name, as this array might be empty or contain an arbitrary number of otherLegalNames.
Thanks.
Regards Artur