Allow ability to set maxDocCharsToAnalyze in WeightedSpanTermExtractor [LUCENE-3287]

Spinoff from this thread:

http://www.gossamer-threads.com/lists/lucene/java-user/129668

In WeightedSpanTermExtractor the default maxDocCharsToAnalyze value is 0. This inhibits us from getting the weighted span terms in any custom code(e.g attached CustomHighlighter.java) that uses WeightedSpanTermExtractor. Currently the setMaxDocCharsToAnalyze() method is protected, which prevents us from setting maxDocCharsToAnalyze to a value greater than 0. Changing the method to public would give us the ability to set the maxDocCharsToAnalyze.

---
Migrated from [LUCENE-3287](https://issues.apache.org/jira/browse/LUCENE-3287) by Jahangir Anwari, updated Jul 11 2011
Attachments: [CustomHighlighter.java](https://apache.github.io/lucene-jira-archive/attachments/LUCENE-3287/CustomHighlighter.java), [WeightedSpanTermExtractor.patch](https://apache.github.io/lucene-jira-archive/attachments/LUCENE-3287/WeightedSpanTermExtractor.patch)

Review WeightedSpanTermExtractor alongside the attached CustomHighlighter.java and WeightedSpanTermExtractor.patch. Start by locating setMaxDocCharsToAnalyze() and confirm how the default value of 0 affects custom callers. Done means callers can set maxDocCharsToAnalyze to a value greater than 0 through the public API.

Written by the indexing model from the issue text.