Enhance Shapefile Reader to handle UTF-8 Code Points

[gwlucastrig]

The Shapefile Reader / DBF file reader should be able to detect code-point files (CPG file extension) and adjust reading to handle non-ASCII cases.

Currently, the reader expect String inputs to be either ASCII or ISO-8859-1 (it doesn't distinguish between them). It should probably be able to handle at least UTF-8 and maybe in ISO-8859-1 mode make sure that it screens out non-valid characters.

Not sure I would bother with all the many variations. Just get these major ones likely to be used by Esri ArcGIS users. Something like Windows 1252 could wait until somebody identifies an actual data set that uses it.

[gwlucastrig]

For abbreviation "cpg", is the correct term "code page" not "code point" ?

A partial update was made for Issue 99, change to DbfField