Highlighter needs the whole text in memory to work [LUCENE-3772]

Highlighter methods getBestFragment(s) and getBestTextFragments only accept a String object representing the whole text to highlight. When dealing with very large docs simultaneously, it can lead to heap consumption problems. It would be better if the API could accept a Reader objetct additionally, like Lucene Document Fields do.

---
Migrated from [LUCENE-3772](https://issues.apache.org/jira/browse/LUCENE-3772) by Luís Filipe Nassif, updated Oct 15 2012
Environment:
```
Windows 7 Enterprise x64, JRE 1.6.0_25
```

Start by locating the Highlighter API methods getBestFragment(s) and getBestTextFragments and tracing how they consume the complete String. Review the related highlighting behavior and tests, then confirm that a Reader-based API can process large documents without requiring the whole text in memory while preserving existing results.

Written by the indexing model from the issue text.