Add option to control whether or not the reader will skip empty rows

[benplante]

This PR addresses #164

Add the IgnoreEmptyRows option to the ExcelDataReaderOptions class. Default behaviour is to skip over any empty rows in the spreadsheet until another non-empty row is found or the spreadsheet ends. In some edge cases a spreadsheet may contain a large number of empty rows which can cause calls to the Read() method to take a long time since it is reading over and skipping many rows. Setting this option to false disables will result in the reader returning every row regardless of how many fields the row contains, which means library users may control if the reader continues to read when a row is empty.

[MarkPflug]

This PR was incomplete, it didn't contain any tests, nor an implementation for .xls files.
I've completed this feature as PR #168. The option is named 'IgnoreEmptyTrailingRows`, because the library never skipped empty rows that have subsequent rows of data, and that behavior is unchanged.