Optimize DAG-CBOR string decode with ASCII fast path

[MarshalX]

No description provided.

[codspeed-hq]

Merging this PR will degrade performance by 12.49%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

❌ 2 regressed benchmarks
✅ 190 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	test_dag_cbor_decode_fixtures[map-1_pair]	11.7 µs	13.3 µs	-12.41%
❌	Simulation	test_dag_cbor_decode_fixtures[map-1_pair_rev]	11.7 µs	13.4 µs	-12.58%

Tip

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

---

_{Comparing optimize-decode-ascii-fast-path (63a1ed3) with main (4c7905a)}

[MarshalX]

CodSpeed does not show it, but locally on M1 it is ~7–12% impovements for decoding github, twitter and citm. No diff for canada (cuz floats)