fix(cucumberjs): use scenario line in testCaseId to prevent duplicate name collision

[todti]

Problem

Fixes #1387.

When two Scenario blocks in the same feature file share an identical name (but have different steps), the reporter computed the same testCaseId for both:

// before
const testCaseId = md5(`${fullNameBase}#${scenario?.name ?? pickle.name}`);

allure-report (the Java/HTML generator) deduplicates test results by testCaseId, so one of the two tests was silently dropped from the report — users saw N-1 tests instead of N. All other reporters (JSON, JUnit XML, HTML) were unaffected because they don't deduplicate.

Fix

Include scenario.location.line in the testCaseId hash. Two scenarios with the same name but on different lines now get distinct ids:

// after
const scenarioLine = scenario?.location?.line ?? "";
const testCaseId = md5(`${fullNameBase}#${scenario?.name ?? pickle.name}#${scenarioLine}`);

Scenario Outline rows are unaffected — all rows of one outline share the same Scenario Outline: line number, so they keep the same testCaseId. This is required for cross-run history tracking (you want to see the history of "row 1 of this outline" across multiple runs, not treat each run as a fresh test).

Breaking Change

testCaseId changes for every test in every feature file, because the hash input gains a #<line> suffix. Existing allure history tied to the old hash will be lost after upgrading — the first run after the upgrade will appear as a new test with no history.

_fallbackTestCaseId (the legacyTestCaseId used for allure2 → allure3 history migration) is unchanged, so the migration path from older allure versions is preserved.

Testing

Added duplicate-names.feature with two Scenario: same name blocks and a new test asserting they produce different testCaseId values. The existing test for Scenario Outline (should create same test case id for scenario outline tests with template name) continues to pass.

[github-actions]

Allure Report Summary

	Name	Duration	Stats	New	Flaky	Retry	Report
	Allure JS – Pull request #1502	2h 28m	4675    14	21	0	0	View