Fix four out-of-bounds defects in the Query serializer

[adwsingh]

What behavior changes?

AWS Query and EC2 Query serialization now:

• Safely handles long Unicode strings and arbitrary-length BigInteger values.
• Preserves outer indexes when serializing nested lists and maps.
• Returns request bodies that are not corrupted when the pooled serializer is reused.
• Avoids re-encoding ASCII prefixes in mixed Unicode strings.
• Transfers oversized buffers instead of copying them before discarding the pooled buffer.

Why is this change needed?

The serializer had several buffer-bound and shared-state defects that could cause out-of-bounds writes, incorrect collection indexes, or request-body corruption. The change also reduces CPU and allocation overhead for large and mixed-Unicode values.

How was this validated?

Integration tests and benchmarks.

What should reviewers focus on?

Point reviewers to the files or sections that contain the interesting logic.

Additional Links

Related issues, design docs, or prior art.

---

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.