fix: escape backslashes and unicode in parametrize test IDs

[hughhan1]

Summary

Fixes #124 - Test IDs for parametrized tests containing escaped unicode strings now match pytest's output.

Previously, strings like "\u2603" (literal backslash-u) would appear as-is in test IDs, but pytest escapes these as "\\u2603". This caused test lookups to fail when using --runner pytest.

The fix adds ascii_escape_string() which escapes:

• Backslashes: \ → \\
• Control characters: \n → \n, \t → \t, etc.
• Non-ASCII: ☃ → \u2603

Test plan

• Unit tests for ascii_escape_string() function
• Integration test comparing rtest vs pytest output
• Manual verification with reproduction case from issue

[github-actions]

Benchmark Comparison Report

Comparing current results against baseline from main branch.

Repository	Benchmark	Baseline	Current	Change	Status
click	CLI startup time	37.6ms	35.3ms	-6.2%	[+]
click	Test discovery performance	47.5ms	46.0ms	-3.2%	[-]
flask	CLI startup time	35.9ms	34.7ms	-3.4%	[-]
flask	Test discovery performance	43.1ms	43.4ms	+0.6%	[-]
more-itertools	CLI startup time	33.8ms	32.1ms	-5.1%	[+]
more-itertools	Test discovery performance	44.3ms	43.9ms	-1.0%	[-]
more-itertools	Test execution with native runner	2.92s	2.16s	-26.0%	[++]
pydantic	CLI startup time	34.1ms	31.3ms	-8.3%	[+]
pydantic	Test discovery performance	140.3ms	134.5ms	-4.1%	[-]

Summary

• Total benchmarks: 9
• Regressions (>5.0% slower): 0
• Improvements (>5.0% faster): 4
• No significant change: 5

Performance improvements detected!

Legend

• [++] Significant improvement (>10.0% faster)
• [+] Minor improvement (5.0-10.0% faster)
• [-] No significant change (±5.0%)
• [!] Minor regression (5.0-10.0% slower)
• [!!] Significant regression (>10.0% slower)