fix: preserve full result description after error ID

[stacknil]

Summary

Fixes #109.

This updates SARIF result record conversion so that when a message starts with the result ruleId, the Description keeps the full remaining message text instead of only the first character after the rule ID.

Root cause

result_to_record() used message[len(error_id) + 1], which indexes a single character. It should slice from that position with message[len(error_id) + 1:].

Validation

• python -m pytest tests/test_sarif_file.py -q
• python -m pytest -q
• python -m ruff check .
• python -m ruff format --check .

Review notes

• Design decision: keep the existing behavior for messages that do not start with the rule ID, and for messages that exactly equal the rule ID; only change the prefixed-message branch from single-character indexing to slicing.
• Main risk: messages that start with the rule ID but use an unusual separator continue to follow the existing prefix/length guard; this PR does not broaden parsing semantics.
• Compatibility impact: non-prefixed and exact rule-ID messages remain unchanged; prefixed messages now preserve the full remaining description instead of one character.
• Rollback path: revert the two commits on this PR branch.

[stacknil]

@microsoft-github-policy-service agree

[stacknil]

Hi maintainers, gentle follow-up on this small fix for #109. I also added boundary coverage for messages that do not start with ruleId and messages that exactly equal ruleId; CLA is green and local pytest/Ruff validation passes. Happy to adjust the test shape if you prefer a different one.