fix(migrations): bring task_records into Alembic, remove guard workaround

- Add d1e2f3a4b5c6 migration: create task_records table matching the
  TaskRecord model (previously created at runtime by init_db() via
  Base.metadata.create_all).
- Update a5b6c7d8e9f0: down_revision now points to d1e2f3a4b5c6, so
  task_records exists when this migration runs. Remove the
  inspector.has_table() guard (no longer needed).
- create_all() in init_db() becomes a no-op for task_records — the
  table is now fully managed by Alembic.

backend/alembic/versions/a5b6c7d8e9f0_set_null_ondelete_for_task_records_env_fk.py

@@ -1,7 +1,12 @@
 """set null ondelete for task_records environment FK
 
+task_records table is now created by prior migration d1e2f3a4b5c6.
+
+This migration alters the FK to add ondelete='SET NULL' for databases
+that were migrated before the FK was defined with ondelete.
+
 Revision ID: a5b6c7d8e9f0
-Revises: c4a3a2f74bfe
+Revises: d1e2f3a4b5c6
 Create Date: 2026-05-21 18:55:00.000000
 
 """
@@ -9,9 +14,10 @@ from collections.abc import Sequence
 
 from alembic import op
 
+
 # revision identifiers, used by Alembic.
 revision: str = 'a5b6c7d8e9f0'
-down_revision: str | Sequence[str] | None = 'c4a3a2f74bfe'
+down_revision: str | Sequence[str] | None = 'd1e2f3a4b5c6'
 branch_labels: str | Sequence[str] | None = None
 depends_on: str | Sequence[str] | None = None
 

backend/alembic/versions/d1e2f3a4b5c6_create_task_records_table.py

@@ -0,0 +1,55 @@
+"""create task_records table
+
+Previously task_records was created at runtime by init_db() →
+Base.metadata.create_all(bind=tasks_engine). This broke Alembic
+migrations that reference task_records (e.g. a5b6c7d8e9f0) — on a
+fresh database, the table didn't exist when migrations ran.
+
+This migration brings task_records into the Alembic-managed schema.
+create_all() becomes a no-op for this table.
+
+Revision ID: d1e2f3a4b5c6
+Revises: c4a3a2f74bfe
+Create Date: 2026-06-11 17:00:00.000000
+
+"""
+from collections.abc import Sequence
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = 'd1e2f3a4b5c6'
+down_revision: str | Sequence[str] | None = 'c4a3a2f74bfe'
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def upgrade() -> None:
+    """Create task_records table matching TaskRecord model."""
+    op.create_table(
+        'task_records',
+        sa.Column('id', sa.String(), nullable=False),
+        sa.Column('type', sa.String(), nullable=False),
+        sa.Column('status', sa.String(), nullable=False),
+        sa.Column('environment_id', sa.String(), nullable=True),
+        sa.Column('started_at', sa.DateTime(timezone=True), nullable=True),
+        sa.Column('finished_at', sa.DateTime(timezone=True), nullable=True),
+        sa.Column('logs', sa.JSON(), nullable=True),
+        sa.Column('error', sa.String(), nullable=True),
+        sa.Column('result', sa.JSON(), nullable=True),
+        sa.Column('created_at', sa.DateTime(timezone=True), server_default=sa.func.now(), nullable=True),
+        sa.Column('params', sa.JSON(), nullable=True),
+        sa.ForeignKeyConstraint(
+            ['environment_id'],
+            ['environments.id'],
+            ondelete='SET NULL',
+            name='task_records_environment_id_fkey',
+        ),
+        sa.PrimaryKeyConstraint('id'),
+    )
+
+
+def downgrade() -> None:
+    """Drop task_records table."""
+    op.drop_table('task_records')