Add ability to process CSV files containing invalid UTF-8 characters

**Is your feature request related to a problem or challenge? Please describe what you are trying to do.**
I have a dataset with invalid UTF-8 characters. Spark is happy to query it, but DataFusion fails with an error. My workaround is to preprocess the file using `String::from_utf8_lossy` but it would be great if I could just set an option in `CsvReadOptions` to have DataFusion do this for me.

**Describe the solution you'd like**
As described.

**Describe alternatives you've considered**
Just preprocess my inputs.

**Additional context**