Since the set of indexed files might be now very large thanks to
SONAR-8623, we should offer to the sensors an optimized way to do the most frequent queries for InputFiles:
- Queries by file name (e.g. replace query by pattern **/pom.xml)
- Queries by file extension (e.g replace query by pattern */.xml)
To do this, we will need to index the input files using a data structure that allows fast queries using those 2 criteria. New FilePredicates should be created to perform the optimized queries.
Bonus: we could try to automatically convert:
- queryByPattern("**/pom.xml") to queryByFilename("pom.xml")
- queryByPattern("*/.xml") to queryByExtension("xml")