Amendment for run_query macro

[xemuliam]

Problem

Different DBT backends treat SQL query differently. For example, BigQuery can run several SQL statements as a single SQL-script.
Snowflake and Databricks can't behave in that way and require to run single SQL statement at a time.
Thus what we can do if we need to run several SQL statements?
Here you can see several options available atb the moment:

• 1st

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

{%- call statement('1') -%}

  select 1

{%- endcall -%}

{%- call statement('2') -%}

  select 2

{%- endcall -%}

select 111

• 2nd

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

{%- set statements = ["
    select 1
  ", "
    select 2
  "
]-%}

{% for stmt in statements %}
  {% do run_query(stmt) %}
{% endfor %}

select 111

• 3rd

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

{%- set full_query %}
  select 1;
  select 2;
  select 3;

{% endset -%}

{%- set statements = full_query.split(';') -%}
{%- for stmt in statements -%}
  {%- if stmt | trim | length > 0 -%}
    {%- do run_query(stmt) -%}
  {%- endif -%}
{%- endfor -%}

select 111

Can we treat that as convenient? Probably not

Solution

Let's amend existing run_query macro to let it be executed in either single statement/script or bulk mode. Default execution mode is script. If someone wants to use it in bulk mode then the only thing needs to be added is a macro's parameter bulk=True.
Proposed change is not breaking. Not even single existing macro invocation will be jeopardized.

Checklist

• I have read the contributing guide and understand what's expected of me
• I have run this code in development and it appears to meet desire functionality
• Specific tests are not required because there is just a wrapper to existing functionality
• This PR has no interface changes (e.g. macros, cli, logs, json artifacts, config files, adapter interface, etc)