Unable to rebuild a Delta model materialized as table

Describe the bug

When using Delta file format, we can run a table model a first time but when trying to rebuild it.

Steps To Reproduce

Create a simple model materialized as table test_table for instance :

{{
    config(
        materialized='table',
        file_format='delta',
    )
}}

with incoming_data as (
    select 1 as id
)

select * from incoming_data

Run this command a first time (no problem) :

dbt run -s test_table

Run it a second time and you will get the following error :

AnalysisException: Cannot create table ('`xxx`.`test_table`'). The associated location ('s3://xxx/xxx/test_table') is not empty but it's not a Delta table

Expected behavior

We should be able to rebuild a table model at any time without issue.

Screenshots and log output

See Steps To Reproduce

System information

The output of dbt --version:

Core:
  - installed: 1.8.3
  - latest:    1.8.4 - Update available!

  Your version of dbt-core is out of date!
  You can find instructions for upgrading here:
  https://docs.getdbt.com/docs/installation

Plugins:
  - glue:  1.8.1 - Up to date!
  - spark: 1.8.0 - Up to date!

The operating system you're using: MacBook Pro 2,3 GHz Intel Core i9 8 cœurs The output of python --version:

Python 3.12.4

Ideas to solve the issue

I think there are multiple possibilities to achieve this :

The easier : add a new materialization delta_table_replace like the already existing iceberge_table_replace
The cleaner (in my opinion) : remove the glue__create_table_as macro and use a new macro (create_glue_table_as for instance) to use in the whole project. This will allow passing on a new parameter to easily be able to pass a should_replace option that adds the "or replace" when called in the table materialization. I would like to hear if there is another reason for keeping the adapter macro since this one will only be used inside the project ? What do you guys think ?

aws-samples / dbt-glue