Add parallel page processing

Enable multi-threaded extraction for multi-page PDFs.

---

## ✅ Challenge Spec

## 🧩 Problem
Identify the performance bottleneck (CPU, memory, OCR time, rendering, parsing hot path, etc.).

## 🎯 Goal
Reduce latency and/or memory usage while preserving extraction correctness.

## ✅ Acceptance Criteria
- Include **before vs after benchmarks** (methodology + environment)
- No functional regressions (tests pass)
- Thread safety validated if adding concurrency
- Any trade-offs documented (accuracy vs speed, memory vs CPU)

## 🧪 Validation
- Benchmark results (numbers + how measured)
- Test/CI evidence
- Notes on JVM flags / environment if relevant

---

---

**Challenge Note:** This issue is part of the ExtractPDF4J Global Build Challenge 2026.
To participate: fork the repo → create a branch → submit a PR referencing this issue → Star the repo.