fix: gate the SIGTERM test to unix and mention shutdown in the quiet docs

[hicksy]

What this changes

Two loose ends from #216. The quiet-mode test stops the server with
kill -TERM, so it's now gated to unix along with its helper: Windows has no
kill, and the server only listens for Ctrl+C there. CI didn't notice because
the Windows job runs --lib only.

The README, docs/http-server.md and the changelog entry said --log quiet
suppressed startup messages. It covers the shutdown line as well, which is
what --help already says.

Checklist

• Tests added or updated

• cargo fmt --check and cargo clippy -- -D warnings pass locally

• CHANGELOG.md updated if this is a user-visible change

• Linked issue, discussion, or a short note explaining the motivation

• I agree my contribution is licensed under the project's terms
  (MIT License and Apache License, Version 2.0)

• DynamoDB compatibility note N/A: nothing here changes an observable
  DynamoDB behaviour.

[github-actions]

Criterion Benchmark Results

Baseline is the per-benchmark median of the last 5 stored runs, so one unusually fast or slow runner cannot skew the comparison. The range column is the spread across those runs.

Benchmark	Baseline (ns/iter)	Range	Current	Change
batch_execute_statement_25	655,396 (n=3)	415,685 - 782,994	823,998	+25.7%
batch_get_item_100	1,404,603	948,986 - 1,451,697	1,506,607	+7.3%
batch_write_item_25	957,722	675,436 - 1,179,405	1,184,406	+23.7%
delete_item	46,152	33,401 - 53,068	53,660	+16.3%
get_item	14,424	9,826 - 15,230	15,843	+9.8%
put_item/put_item/large	269,725	145,467 - 278,302	263,352	-2.4%
put_item/put_item/medium	36,564	25,477 - 46,751	48,899	+33.7%
put_item/put_item/small	20,582	12,922 - 26,885	27,907	+35.6%
query_base_table	1,159,394	820,961 - 1,204,548	1,192,371	+2.8%
query_gsi	22,277	15,019 - 25,843	27,368	+22.9%
scan_with_filter	8,730,220	6,103,566 - 9,113,670	8,964,152	+2.7%
transact_write_items_4	224,380	157,077 - 272,395	283,197	+26.2%
update_item	163,483	146,389 - 170,240	151,465	-7.4%

All benchmarks within 50% of the 5-run median.

Runs in the baseline

• runs/2026-07-30-bc2a16c
• runs/2026-07-30T220642Z-be8bfbc
• runs/2026-08-25T200503Z-54a2cb6
• runs/2026-09-20T153257Z-49c6338
• runs/2026-09-20T154723Z-49c6338