maybe yes, we are thinking about it (also for Redis, MongoDB, PostgreSQL)

But in the meantime you can use external mongodb providers like MongoDB Atlas:
https://www.mongodb.com/cloud/atlas
https://www.mongodb.com/pricing

They even provide a free tier for small databases

you can they use it from your cloud functions with the NPM package: https://www.npmjs.com/package/mongodb

const MongoClient = require('mongodb').MongoClient;
const uri = "mongodb+srv://root:<password>@XXXXXXXXX.mongodb.net/<dbname>?retryWrites=true&w=majority";
const client = new MongoClient(uri, { useNewUrlParser: true });
client.connect(err => {
  const collection = client.db("test").collection("devices");
  // perform actions on the collection object
  client.close();
});

Indeed we don't expose Point (PT) type for now, this may be added later

In the meantime you can use "New Query" in the left menu to add your column with SQL:
https://dev.mysql.com/doc/refman/8.0/en/creating-spatial-columns.html

Here is a nice representation of the precision you'll have depending your choice : http://mysql.rjweb.org/doc.php/latlng
float would be 1.7m, so it all depends what you want to do with it.

Ideally you should the POINT from this : https://dev.mysql.com/doc/refman/8.0/en/spatial-types.html
But I don't think it's available on AppDrag ? I'll let them answer on that one ^^"

Good day