Reducing GraphQL schema size

[ikenox]

Hello, I'm using payloadcms for my current project, and it's a really helpful to develop our application rapidly.

However, I've faced a problem that the generated GraphQL schema is quite large.
I currently have only 10+ collections in my project, but the resulting GraphQL schema exceeds 20k+ lines.

Because of the large schema, it takes 5-10 minutes for auto-generating the our client code (flutter project). It also results in a large amount of generated code that consumes memory resource on editor.

Upon examining the schema, I noticed that it could potentially be reduced in size with a few simple adjustments.
For instance, for every unique String field, almost the same filter condition input types are defined.
Here is an example:

input ExampleCollection_where {
  field1: ExampleCollection_field1_operator
  field2: ExampleCollection_field2_operator
  field3: ExampleCollection_field3_operator
  updatedAt: ExampleCollection_updatedAt_operator
  createdAt: ExampleCollection_createdAt_operator
  email: ExampleCollection_email_operator
  id: ExampleCollection_id_operator
  AND: [ExampleCollection_where_and]
  OR: [ExampleCollection_where_or]
}

input ExampleCollection_field1_operator {
  equals: String
  not_equals: String
  like: String
  contains: String
  in: [String]
  not_in: [String]
  all: [String]
  exists: Boolean
}

input ExampleCollection_field2_operator {
  equals: String
  not_equals: String
  like: String
  contains: String
  in: [String]
  not_in: [String]
  all: [String]
  exists: Boolean
}

input ExampleCollection_field3_operator {
  equals: String
  not_equals: String
  like: String
  contains: String
  in: [String]
  not_in: [String]
  all: [String]
  exists: Boolean
}

Using the same input type for filters of the same type seems likely to reduce the schema size.

In addition to this, it would be great to have an option that allows to exclude specific Queries and Mutations from the schema.
We do not plan to use queries like docAccessXxx that are generated for each collections, so excluding them would potentially lead to a significant reduction in schema size.

I would appreciate any thoughts or feedback. Thanks!