fix: stabilize best-first batch ordering

[nightcityblade]

Summary

Fixes #1998 by making best-first batch processing deterministic even when streamed crawl results complete out of order.

List of files changed and why

• crawl4ai/deep_crawling/bff_strategy.py - collect concurrent batch results first, then process/yield them in priority-queue order before discovering links.
• tests/deep_crawling/test_deep_crawl_resume.py - add a regression test for out-of-order streamed batch results.

How Has This Been Tested?

• pytest tests/deep_crawling/test_deep_crawl_resume.py::TestBestFirstResume::test_stream_results_follow_priority_order_with_out_of_order_batch -q
• pytest tests/deep_crawling/test_deep_crawl_resume.py::TestBestFirstResume -q

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have added/updated unit tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes