release: 0.17.0

[samuelcolvin]

• Bump the version to 0.17.0 (Cargo.toml, Cargo.lock).
• Regenerate the README benchmark table with make bench-table at JITER_BENCH_SEED=0 and the json-cases corpus at the CI-pinned commit. Every row has moved since the last refresh, from the SIMD string scanning, the digit-run number decoder, the shared container stacks and the string cache changes; json_cases_all goes from 26.7ms to 22.2ms.
• Register doubles_array, short_floats and long_significand_floats (added in Float tests and per-shape benchmarks #283) in the table script's numbers group, so they stop falling into the "other" bucket.
• Add mdformat to the linting group with make format-md / make lint-md and a pre-commit hook that runs the formatter. mdformat's defaults are the style we want, so there's no config file. This is what padded and aligned the table's columns; the only other change it made is * to - bullets.

🤖 Generated with Claude Code

https://claude.ai/code/session_01Qm4KCoN9xWsA4HRCL3Lx63

---

Summary by cubic

Bumps the version to 0.17.0 and regenerates the benchmark table, which shows broad gains from the SIMD string scanning, digit-run number decoder, shared container stacks, and string cache changes — json_cases_all drops from 26.7ms to 22.2ms.

• Registers doubles_array, short_floats, and long_significand_floats (added in Float tests and per-shape benchmarks #283) in the table script's numbers group so they no longer fall into the "other" bucket.
• Adds mdformat and mdformat-tables to the linting group, exposed via make format-md / make lint-md and a pre-commit hook; the formatter realigned the table columns and converted * bullets to -.
• The markdown recipes target README.md and crates explicitly (excluding hidden and target dirs) since mdformat ignores .gitignore.
• Ruff and markdown recipes run with uv run --group linting so they work in fresh checkouts where uv syncs only the default dev group.

^{Written for commit 6782b5d. Summary will update on new commits.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[cubic-dev-ai]

All reported issues were addressed

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>

[codspeed]

Merging this PR will degrade performance by 6.86%

⚡ 1 improved benchmark
❌ 2 regressed benchmarks
✅ 64 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
❌	unicode_jiter_iter	1 µs	1.3 µs	-17.93%
❌	sentence_jiter_iter	1 µs	1.1 µs	-8.24%
⚡	string_array_jiter_value_owned	13.8 µs	12.9 µs	+7.29%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing release-0.17.0 (6782b5d) with main (6881310)}

[cubic-dev-ai]

All reported issues were addressed across 2 files (changes from recent commits).

<sub>Tip: Review your code locally with the cubic CLI to iterate faster.

Re-trigger cubic</sub>