Use ConcurrentHashMap for DocumentContainer parser registries

[rootvector2]

Thanks for your contribution to Apache Commons!

• Read the contribution guidelines for this project.
• Read the ASF Generative Tooling Guidance if you use Artificial Intelligence (AI).
• I used AI to create any part of, or all of, this pull request. Which AI tool was used to create this pull request, and to what extent did it contribute?
• Run a successful build using the default Maven goal with mvn; that's mvn on the command line by itself.
• Write unit tests that match behavioral changes, where the tests fail if the changes to the runtime are not applied.
• Write a pull request description that is detailed enough to understand what the pull request does, how, and why.
• Each commit in the pull request should have a meaningful subject line and body.

DocumentContainer keeps two static registries, parsers and parserClasses, as plain HashMap. registerXMLParser mutates them with put, and getParser (reached through getValue then parseXML) mutates parsers with computeIfAbsent, none of it synchronized, while the rest of the library already uses ConcurrentHashMap for the same kind of registry (JXPathIntrospector, ValueUtils). When two threads first parse a given model concurrently, or one thread registers a parser while another parses, they race on the shared map. I hit a ConcurrentModificationException thrown from HashMap.computeIfAbsent in getParser, and on other runs a corrupted map spinning forever in resize/treeify inside registerXMLParser at 100% CPU. Switching both fields to ConcurrentHashMap lines them up with the other registries and drops the race. The added test reproduces it: it fails on the current code with the ConcurrentModificationException and passes with the change.

[garydgregory]

@rootvector2
Merged, thank you.