Auto-detect the character encoding set of datasets on import

[clayliddell]

Currently, DKAN only supports UTF-8 datasets. MySQL allows defining character set on a table by table basis, so we should be able to support datasets of any character encoding set, we'll just need to devise some mechanism for detecting a dataset file's encoding.

User Story

As a DKAN user, I want to be able to upload and import non-UTF-8 encoded CSV files.

Acceptance Criteria

When a non-UTF-8 encoded CSV file is uploaded, it is properly imported and displayed on a dataset.