test: close the audit coverage gaps

[MaxMichel2]

Summary

Implements PR 11 from the NetworkMock backlog handoff: closes the remaining test-coverage gaps tracked in #91.

Per the handoff doc's instruction, I re-scoped each of #91's seven gaps against current code before writing anything — the issue predates the OpenAPI migration and references files/classes that no longer exist (NetworkMockEndpointScreen → OperationPickerPage, NetworkMockEndpointPreviewBottomSheet.kt → MockResponsePreviewPage.kt).

Already covered (verified, not duplicated)

Gap	Status
Ambiguous host-match precedence	MockConfigRepositoryTest's findMatchingMock picks the first spec that has a matching operation when hosts collide / falls through to the next spec...
Preview/diff bottom sheet	MockResponsePreviewPageTest.kt (page) + DiffLineUtilsTest.kt (diff logic)
Response Content-Type/header assertions	Closed by #127 (returnsMockResponse_withDefaultContentTypeHeader / _withDeclaredHeadersAndContentType)

Genuinely open — closed by this PR

1. Delay precedence — existing test covered operation-override and spec-default fallback, but not the "neither declares one → null" case. Added one test to complete the chain.
2. Real sample spec parsing (devview-networkmock-core, new RealSampleSpecTest, androidHostTest) — no test previously loaded the actual shipped specs (sample/network's sample-api.json, jsonplaceholder.json) through the real MockConfigRepository; every other test uses hand-built inline fixtures, so the shipped sample could silently drift out of sync with what the parser accepts. Reads the files directly off disk via a devview.sampleNetworkResourcesDir system property (computed once at Gradle configuration time via rootProject.file(...), so it doesn't depend on the JVM working directory) — deliberately not a compile-time dependency on sample:network, which would invert this module's place in the dependency graph.
3. Query-parameter matching end-to-end through the Ktor plugin — RequestMatcherTest already covered matchesQueryParams in isolation, but NetworkMockPluginTest had no coverage through the actual interception path. KtorPluginTestData's shared spec gains a listUsers operation (?type=user, mirroring the real sample spec), and NetworkMockPluginTest gains a "Query parameter matching" region (3 tests: matches, wrong value falls through, missing param falls through).
4. Sticky-header status-family grouping — OperationPickerPage already groups responses by StatusCodeFamily with a sticky header per group, but nothing asserted the header text or that responses land in distinct groups. NetworkMockOperationSheetTest's existing fixture already spans two families (2xx/4xx); one added test asserts both headers render.

Testing

• MockConfigRepositoryTest: +1 test (delay precedence completion).
• RealSampleSpecTest (new, androidHostTest): 4 tests — both real sample specs parse, and a representative operation's declared response files all load.
• NetworkMockPluginTest: +3 tests (query-parameter matching region).
• NetworkMockOperationSheetTest: +1 test (sticky-header grouping).
• CHANGELOG.md documents this as a coverage-closing entry (no user-facing behavior change).

Verification run locally

• :devview-networkmock-core:testAndroidHostTest (commonTest + androidHostTest, incl. new RealSampleSpecTest) — ✅
• :devview-networkmock-ktor:testAndroidHostTest — ✅
• :devview-networkmock:compileAndroidDeviceTest — ✅
• detektFull — ✅
• :konsist:test — ✅
• :sample:androidApp:assembleDebug — ✅

Closes #91.