I think for your purposes the cygwin diff utility should work fine (you can use it without needing to install cygwin). If you need finer grained comparison, you might be able to build on CSV table diff utility.

In regards to using ODBC to do the searching - if your files are large, you may gain some speed that way, but trying it yourself is the only way to know how much you'd gain.