fix(config): make default seed deterministic

[gargiulofrancesco]

Makes DEFAULT_CONFIG deterministic by changing its seed from None to Some(42). This ensures reproducible dictionary training with the default settings.

[codspeed-hq]

Merging this PR will improve performance by 56.44%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ 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

⚡ 12 improved benchmarks
✅ 20 untouched benchmarks
⏩ 2 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	WallTime	decompress_all[12]	750.2 µs	312.1 µs	×2.4
⚡	WallTime	decompress_all[16]	746.5 µs	323.9 µs	×2.3
⚡	WallTime	decompress_all[("p_name", 12)]	1,121.2 µs	502.8 µs	×2.2
⚡	WallTime	decompress_all[("p_name", 16)]	912.3 µs	501.1 µs	+82.06%
⚡	WallTime	decompress_all[("l_comment", 12)]	16 ms	9 ms	+77.36%
⚡	WallTime	decompress_all[("o_comment", 12)]	14.3 ms	8.2 ms	+73.51%
⚡	WallTime	decompress_all[("o_comment", 16)]	13.6 ms	9.9 ms	+36.79%
⚡	WallTime	decompress_all[("l_comment", 16)]	15 ms	11.4 ms	+31.62%
⚡	WallTime	train_and_compress[("p_name", 16)]	39.4 ms	33.7 ms	+16.88%
⚡	WallTime	train_and_compress[("o_comment", 16)]	474.8 ms	412.3 ms	+15.16%
⚡	WallTime	train_and_compress[16]	35.2 ms	30.9 ms	+14.07%
⚡	WallTime	train_and_compress[12]	31.5 ms	28.1 ms	+12.35%

Tip

Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing fix/deterministic-default-seed (7bf548b) with develop (a31dfd5)}

Footnotes

1. 2 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[0ax1]

No CI for the on-pair repo? cc: @joseph-isaacs @gargiulofrancesco Or rather why are the tasks not visible in an overview here?