Closing the DirTaxonomyReader while another thread access the cache can throw NPE [LUCENE-10077]

When we close a `DirectoryTaxonomyReader` in `doClose`, we set the `categoryCache` to null. But if a thread is next after this `doClose` call, it will still try to acquire a lock and `synchronize` on it. This will result in an NPE.

This works well today, because we operate on the assumption that the user will always call `doClose` after all threads have completed.
One suggestion by @mikemccand in this [PR](https://github.com/apache/lucene/pull/179#discussion_r697880516) was to make categoryCache final and throw an AlreadyClosedException.

---
Migrated from [LUCENE-10077](https://issues.apache.org/jira/browse/LUCENE-10077) by Gautam Worah (@gautamworah96), updated Sep 01 2021

Start at DirectoryTaxonomyReader#doClose and trace how categoryCache is accessed during close. Reproduce or cover the concurrent access case, then verify that a reader closed during cache access no longer throws an unexpected NullPointerException and instead follows the intended closed-reader behavior.

Written by the indexing model from the issue text.