npgsql: Closing connection before reader disposing leads to broken connector

Steps to reproduce

MyMethod()
{
            if (_connection.State != ConnectionState.Open)
            {
                await _connection.OpenAsync(token);
                await _command.PrepareAsync(token);
            }

            using var reader = await _command.ExecuteReaderAsync(token);

            while (await reader.ReadAsync(token))
            {
                ...
                    yield return someResult;
                    
            }

            if (_connection.State != ConnectionState.Closed)
                _connection.Close();

            if (...)
                yield return someResult;
}

The issue

I have 2 Exceptions one by one:

1. For new opening connection (exception messages can be different)

Npgsql.NpgsqlException (0x80004005): Received backend message BindComplete while expecting ReadyForQueryMessage. Please file a bug.\n  
 at Npgsql.NpgsqlDataReader.NextResult(Boolean async, Boolean isConsuming)\n  
 at Npgsql.NpgsqlCommand.ExecuteReaderAsync(CommandBehavior behavior, Boolean async, CancellationToken cancellationToken)\n   
at Npgsql.NpgsqlCommand.ExecuteNonQuery(Boolean async, CancellationToken cancellationToken)\n  
 at Lib.Pgsql.ConnectionExtensions.OpenWithParametersAsync(NpgsqlConnection connection, String defaultSchema)\n  
 at ...

1. For reader disposing

Npgsql.NpgsqlException (0x80004005): Exception while reading from stream\n ---> System.IO.IOException: Unable to read data from the transport connection: Operation canceled.\n ---> System.Net.Sockets.SocketException (125): Operation canceled\n   
at System.Net.Sockets.NetworkStream.Read(Byte[] buffer, Int32 offset, Int32 size)\n  
 --- End of inner exception stack trace ---\n   
at System.Net.Sockets.NetworkStream.Read(Byte[] buffer, Int32 offset, Int32 size)\n   
at Npgsql.NpgsqlReadBuffer.<>c__DisplayClass34_0.<<Ensure>g__EnsureLong|0>d.MoveNext()\n  
at Npgsql.NpgsqlReadBuffer.<>c__DisplayClass34_0.<<Ensure>g__EnsureLong|0>d.MoveNext()\n
--- End of stack trace from previous location where exception was thrown ---\n  
 at Npgsql.NpgsqlConnector.<>c__DisplayClass160_0.<<DoReadMessage>g__ReadMessageLong|0>d.MoveNext()\n
--- End of stack trace from previous location where exception was thrown ---\n  
 at Npgsql.NpgsqlConnector.<>c__DisplayClass160_0.<<DoReadMessage>g__ReadMessageLong|0>d.MoveNext()\n
--- End of stack trace from previous location where exception was thrown ---\n 
  at Npgsql.NpgsqlDataReader.NextResult(Boolean async, Boolean isConsuming)\n  
 at Npgsql.NpgsqlDataReader.Consume(Boolean async)\n  
 at Npgsql.NpgsqlDataReader.Close(Boolean connectionClosing, Boolean async)\n 
  at Npgsql.NpgsqlDataReader.Close()\n   at 
SomeClass.MyMethod(...)   at 

And finally this leads to exhausted connection pool. I got my mistake in the code above but I hope you can make this issue more explicit.

Further technical details

Npgsql version: 4.1.3 PostgreSQL version: 10.7 Operating system: Any