Stop Words Should Not Be Hard Coded

During text index creation, we pass in a hard coded list of stop words which limits the vocabulary that we can use to filter words before indexing.

We should use Lucene's StopFilter and read the list of stop words from a file, thus allowing users flexibility in terms of filtering words.

Start by locating the text index creation path and the hard-coded stop-word list described in the issue. Read how Lucene's StopFilter is used there and identify the existing configuration or resource-loading conventions; done means users can provide a stop-word file that is read during indexing instead of relying on the fixed list.

Written by the indexing model from the issue text.