Build TABLESWITCH from sorted arrays in SWITCH

[rootvector2]

SWITCH clones the caller's match/targets arrays and sorts only the clones, but the TABLESWITCH branch fills the table from the original unsorted match/targets, while the LOOKUPSWITCH branch uses matchClone/targetsClone. When the match array is passed unsorted (the class javadoc documents unsorted input and promises to leave the caller's arrays unaltered) and its sorted form is gap-free, the fill loop runs over unsorted keys, so the generated TABLESWITCH gets out-of-order case values padded with default entries. TABLESWITCH.dump then writes low = match[0] and high = match[last] that disagree with the count of offset words emitted, and re-reading the dumped instruction consumes only high - low + 1 offsets, reinterpreting the surplus words as the next instructions.

Fill the table from matchClone/targetsClone to match the LOOKUPSWITCH branch. Found while auditing the switch-generation paths.

• 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. This may not always be possible, but it is a best practice.
• 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. Note that a maintainer may squash commits during the merge process.