Include the regex pattern in SubfieldPatternQuery's repr

[dchud]

repr(SubfieldPatternQuery(...)) showed <SubfieldPatternQuery tag=084 subfield=a>, omitting the regex — the most useful field for debugging a query at the REPL or in logs. Now:

<SubfieldPatternQuery tag=084 subfield=a pattern="^abc">
<SubfieldPatternQuery tag=100 subfield=d pattern="\\d{4}-\\d{4}" negate=true>

Changes across all four layers:

• Rust core — SubfieldPatternQuery::pattern() -> &str (via Regex::as_str(), no extra storage).
• PyO3 — a pattern getter, and the pattern included in both __repr__ arms. The pattern is debug-quoted ({:?}) so a value with spaces or special characters stays legible and a pattern containing a quote can't break the repr — the tradeoff is that backslashes show escaped (\\d).
• Stubs — pattern declared in _mrrc.pyi, along with the tag / subfield_code getters that were already exposed in PyO3 but missing from the stub.
• Tests — Rust accessor test; Python repr + getter assertions for both negated and non-negated queries.

.cargo/check.sh green (566 Rust tests + doc tests + clippy + ruff + mkdocs). CHANGELOG [Unreleased] credits @acdha.

Reported by @acdha in #226.

Closes #235

Bead: bd-h9v4

[codspeed-hq]

Merging this PR will degrade performance by 36.85%

⚠️ 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

⚡ 5 improved benchmarks
❌ 47 (👁 47) regressed benchmarks
✅ 8 untouched benchmarks
⏩ 18 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
👁	WallTime	test_file_sequential_4x_10k	233.4 ms	395.7 ms	-41.01%
👁	WallTime	test_memory_field_creation_bulk	45.4 ms	58.7 ms	-22.59%
👁	WallTime	test_read_and_extract_titles_10k	55.7 ms	84.3 ms	-33.95%
👁	WallTime	test_memory_read_10k_records	199.2 ms	230.2 ms	-13.44%
👁	WallTime	test_write_only_1k_rustfile	3.5 ms	4.3 ms	-17.75%
👁	WallTime	test_collect_all_records_1k	5.4 ms	8 ms	-32.75%
👁	WallTime	test_stream_write_1k	8.1 ms	13.3 ms	-38.83%
👁	WallTime	test_read_and_extract_titles_1k	5.4 ms	8.3 ms	-35.41%
👁	WallTime	test_sequential_reading_1k	5.7 ms	10.9 ms	-47.55%
👁	WallTime	test_memory_streaming_read_10k	212.6 ms	248.5 ms	-14.47%
👁	WallTime	test_sequential_2x_reading_1k	10.5 ms	16.7 ms	-37.17%
👁	WallTime	test_sequential_4x_reading_1k	20.1 ms	33.3 ms	-39.73%
👁	WallTime	test_threaded_reading_4x_1k	47 ms	109 ms	-56.9%
👁	WallTime	test_sequential_10k	53.6 ms	84 ms	-36.2%
👁	WallTime	test_threaded_reading_1k	15.4 ms	30.8 ms	-49.92%
⚡	WallTime	test_pipeline_parallel_4x_10k_threaded	125.9 ms	109.8 ms	+14.64%
👁	WallTime	test_sequential_2x_reading_10k	108.9 ms	168.9 ms	-35.53%
⚡	WallTime	test_pipeline_parallel_extraction_4x_10k_threaded	149 ms	134 ms	+11.15%
👁	WallTime	test_threading_speedup_2x_10k	161.2 ms	311.6 ms	-48.26%
⚡	WallTime	test_process_4_files_sequential	127.9 ms	93.1 ms	+37.29%
...	...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

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 feat/subfield-pattern-repr (bb045f4) with main (df26a26)}

Footnotes

1. 18 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. ↩