fix(docker): preserve valid seeds in crawl batches - #2347

Summary

Fixes #2288.

Mixed /crawl batches now keep destination validation per seed: blocked or unresolvable seeds become opaque failed results, while valid seeds are still crawled. Results and per-URL crawler configs retain their original input order. Single-URL requests keep the existing HTTP 400 behavior.

List of files changed and why

• deploy/docker/api.py - Partition multi-URL batches into valid seeds and per-seed validation failures, crawl the valid subset, and merge results in request order.
• tests/test_docker_pdf_crawler_pairing.py - Add regression coverage for one blocked seed alongside one valid seed.

How Has This Been Tested?

• uv sync --python 3.12 --frozen --dev
• uv pip install --python .venv/bin/python -r deploy/docker/requirements.txt -r deploy/docker/tests/requirements.txt pytest pytest-asyncio black ruff
• .venv/bin/python -m pytest deploy/docker/tests/test_security_ssrf_crawl.py tests/test_docker_pdf_crawler_pairing.py -q — 37 passed
• .venv/bin/ruff check --select E9,F63,F7,F82 deploy/docker/api.py tests/test_docker_pdf_crawler_pairing.py — passed
• .venv/bin/python -m compileall -q deploy/docker/api.py tests/test_docker_pdf_crawler_pairing.py — passed
• Manual async handler regression with one blocked and one valid seed — passed; only the valid URL reached arun_many, and both ordered results were returned
• git diff --check upstream/develop...HEAD — passed

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 — N/A; no public API or configuration changed
• 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