Fix (or silence?) occasional postgrex connection error in tests

This doesn't result in test failures, but all the red text could be distracting or misleading to devs debugging other failing tests: 07:59:05.968 [error] Postgrex.Protocol (#PID<0.1793.0>) disconnected: ** (DBConnection.ConnectionError) client #PID<0.3956.0> exited

The logs are:

07:59:09.327 [error] Postgrex.Protocol (#PID<0.1798.0>) disconnected: ** (DBConnection.ConnectionError) owner #PID<0.5222.0> exited

Client #PID<0.5268.0> is still using a connection from owner at location:

    :prim_inet.recv0/3
    (postgrex 0.16.5) lib/postgrex/protocol.ex:3171: Postgrex.Protocol.msg_recv/4
    (postgrex 0.16.5) lib/postgrex/protocol.ex:2871: Postgrex.Protocol.recv_transaction/4
    (postgrex 0.16.5) lib/postgrex/protocol.ex:2180: Postgrex.Protocol.rebind_execute/4
    (ecto_sql 3.9.0) lib/ecto/adapters/sql/sandbox.ex:375: Ecto.Adapters.SQL.Sandbox.Connection.proxy/3
    (db_connection 2.4.2) lib/db_connection/holder.ex:354: DBConnection.Holder.holder_apply/4
    (db_connection 2.4.2) lib/db_connection.ex:1364: DBConnection.run_execute/5
    (db_connection 2.4.2) lib/db_connection.ex:1459: DBConnection.run/6
    (db_connection 2.4.2) lib/db_connection.ex:652: DBConnection.execute/4
    (ecto_sql 3.9.0) lib/ecto/adapters/postgres/connection.ex:102: Ecto.Adapters.Postgres.Connection.execute/4
    (ecto_sql 3.9.0) lib/ecto/adapters/sql.ex:850: Ecto.Adapters.SQL.execute!/5
    (ecto_sql 3.9.0) lib/ecto/adapters/sql.ex:820: Ecto.Adapters.SQL.execute/6
    (ecto 3.9.1) lib/ecto/repo/queryable.ex:229: Ecto.Repo.Queryable.execute/4
    (ecto 3.9.1) lib/ecto/repo/queryable.ex:19: Ecto.Repo.Queryable.all/3
    (ecto 3.9.1) lib/ecto/repo/preloader.ex:272: Ecto.Repo.Preloader.fetch_query/8
    (elixir 1.14.1) lib/task/supervised.ex:89: Task.Supervised.invoke_mfa/2
    (elixir 1.14.1) lib/task/supervised.ex:34: Task.Supervised.reply/4
    (stdlib 4.1.1) proc_lib.erl:240: :proc_lib.init_p_do_apply/3

The connection itself was checked out by #PID<0.5268.0> at location:

    (ecto_sql 3.9.0) lib/ecto/adapters/postgres/connection.ex:102: Ecto.Adapters.Postgres.Connection.execute/4
    (ecto_sql 3.9.0) lib/ecto/adapters/sql.ex:850: Ecto.Adapters.SQL.execute!/5
    (ecto_sql 3.9.0) lib/ecto/adapters/sql.ex:820: Ecto.Adapters.SQL.execute/6
    (ecto 3.9.1) lib/ecto/repo/queryable.ex:229: Ecto.Repo.Queryable.execute/4
    (ecto 3.9.1) lib/ecto/repo/queryable.ex:19: Ecto.Repo.Queryable.all/3
    (ecto 3.9.1) lib/ecto/repo/preloader.ex:272: Ecto.Repo.Preloader.fetch_query/8
    (elixir 1.14.1) lib/task/supervised.ex:89: Task.Supervised.invoke_mfa/2
    (elixir 1.14.1) lib/task/supervised.ex:34: Task.Supervised.reply/4
    (stdlib 4.1.1) proc_lib.erl:240: :proc_lib.init_p_do_apply/3

Start by running the test suite and tracing the disconnect through Ecto.Adapters.SQL.Sandbox.Connection.proxy/3 and Postgrex.Protocol.msg_recv/4, the entry points shown in the logs. Inspect the test database setup to determine why the owner or client exits during a query. Done means the intermittent red Postgrex errors are addressed or silenced without hiding genuine test failures.