Tmp Schemas need to account for Location/Region when orchestrating BigQuery

[ReadytoRocc]

Describe the bug
I am unable to run queries using the astro.sql.transform decorator when they do not exist in the region where tmp_astro is created. In the below example error, I am running a query on a table in us-central1, when tmp_astro exists in europe-west2:

sqlalchemy.exc.DatabaseError: (google.cloud.bigquery.dbapi.exceptions.DatabaseError) 404 Not found: Dataset PROJECT_ID:tmp_astro was not found in location us-central1

Version

• Astro: 1.3.0
• OS: Debian

To Reproduce
Steps to reproduce the behavior:

• Run the below DAG on using a TABLE_PATH in one region.
• Then rerun it again using a different table path in another region.

"""
gcp_bq
DAG auto-generated by Astro Build.
"""

from airflow.decorators import dag
from astro import sql as aql
from astro.sql.table import Table
import pendulum


@aql.transform(conn_id="bq_fe_sa", task_id="cell_2")
def cell_2_func():
    return """SELECT * FROM TABLE_PATH LIMIT 10;"""

@dag(
    schedule_interval="0 0 * * *",
    start_date=pendulum.from_format("2022-12-02", "YYYY-MM-DD").in_tz("America/Phoenix"),
)
def gcp_bq():
    cell_2 = cell_2_func()

dag_obj = gcp_bq()

Expected behavior
I would expect the Astro SDK to make a tmp_astro dataset per region, and ensure that it is using/refencing the appropriate tmp schema based on the source table in the query.

Additional context
The BigQueryHook supports a location parameter, allowing you to specify the location/region for your query. I don't believe this can be set via a Connection parameter.