Getting distinct documents in aggregate pipeline in MongoDB

Question

I'm trying to do union between an Atlas text search and a match in Mongodb, an then get the distinct results from those before adding additional stages. I'm close, but I'm missing a step here I think. There is where I'm at now.

db.getCollection("MyCollection").aggregate([
{
    $search: {
        "phrase": {
            "query": "searchquery",
            "path": "field to look in"
        }
    }
},
{
    $unionWith: {
        coll: "MyCollection",
        pipeline: [{              
            $match: { "someproperty": { $in : [arrayvalues] } }
      }]
  }

} ])

This works fine. If either the search or the match finds a document, it gets added to the documents in the aggregate function. The problem is if the document is found in both, then it appears here twice.

I found no proper way to add a way of getting the distinct list of document here. It seems we are supposed to use $group, and it does almost what I want it to do.

I added the following stage after the $unionWith stage:

{
    $group: { _id: "$_id", doc: { $first: "$$ROOT" } }
}

This ends up removing the duplicates, but I'm now stuck with a result where I have the documents not as they originally was. Now they look like { _id: <distinctId>, doc: { all fields from document} }

Basically I want that last step to give me the first document of each group, but not wrapped inside a new JSON object.