feat(connectors): BI-6992 Fix postgres ARRAY type detection

[catsona]

No description provided.

[Copilot]

Pull request overview

Fixes PostgreSQL connector metadata inference for ARRAY columns by correcting OID-to-type mappings and adding regression tests to validate BI type detection for common array element types.

Changes:

• Corrected PostgreSQL pg_type.oid mappings for _text and _numeric arrays in the adapter.
• Added API-level DashSQL tests validating BI type detection for integer, float/numeric, and string array columns (both table-based and query-based).

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
lib/dl_connector_postgresql/dl_connector_postgresql/core/postgresql_base/adapters_base_postgres.py	Fixes OID mappings used to infer SQLAlchemy/native types for PostgreSQL array columns.
lib/dl_connector_postgresql/dl_connector_postgresql_tests/db/api/test_array_types.py	Adds regression tests ensuring array BI types are detected correctly from metadata.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[KonstantAnxiety]

What do these tests have to do with dashsql? Don't these tests look more suitable for type recognition?

datalens-backend/lib/dl_connector_postgresql/dl_connector_postgresql_tests/db/core/test_data_source.py

Line 54 in 19078d7

class TestPostgreSQLSubselectDataSource(

[catsona]

I've implemented similar to YDB connector, where type recognition is tested using DashSQL:

datalens-backend/lib/dl_connector_ydb/dl_connector_ydb_tests/db/api/test_dashsql.py

Line 13 in 141ddae

async def test_result(

[KonstantAnxiety]

The original issue is related to SQL-datasource, not dashsql

[catsona]

moved test to type recognition