Use column name instead of programatical name about domino-jna

[klehmann]

I rewrote the project code to always use the programmatic names because that's what is returned in the summary buffer when READ_MASK_SUMMARY is used. In general it is better to use READ_MASK_SUMMARYVALUES instead, because more data fits into the 64 K summary buffer (contains values only and no duplicate column names), but for NIFFindByKeyExtended2 in 9.0.1, only READ_MASK_SUMMARY is supported.

With READ_MASK_SUMMARYVALUES is used, I first scan the columns (https://github.com/klehmann/domino-jna/blob/master/domino-jna/src/main/java/com/mindoo/domino/jna/NotesCollection.java#L1943) and read the programmatic names using the legacy Java API (haven't written the code to do this in C yet) to be able to unify both API calls.

It would be possible to always scan the columns and translate the column name argument to the right index (for READ_MASK_SUMMARYVALUES) / programmatic name (for READ_MASK_SUMMARY). I have that on my list, because I want to use the Domino JNA for a project where I am currently using column names instead of programmatic names. But I haven't had the time to work on this.