fix(engine): weight local-scorer totalTokenScore with TEST_FILE_CONTRIBUTION_WEIGHT

[philluiz2323]

Problem

Closes #8875.

computeLocalScorerTokens returned a raw unweighted line sum as totalTokenScore. When that value is fed back into buildScorePreview as an explicit total, preview honors it as-is and skips the TEST_FILE_CONTRIBUTION_WEIGHT (0.05×) discount it applies when deriving its own total — so a test-heavy local scorer preview over-counts test lines in the contribution-bonus ramp.

Fix

Compute totalTokenScore as source + TEST_FILE_CONTRIBUTION_WEIGHT * test + nonCode, matching preview.ts's derivedTotalTokenScore, using the shared DEFAULT_SCORING_CONSTANTS constant.

Tests

• Existing classification fixtures updated to expect the weighted total.
• New test-heavy fixture: local-scorer total equals the weighted formula, and feeding that total into buildScorePreview matches the derived-total (no explicit total) path for contribution bonus / estimated merged score.
• MCP loopover_run_local_scorer expectation updated accordingly.

Reverting the weight fails the agreement test. git diff --check clean.

Scope

• Conventional Commit title
• Focused change
• Follows CONTRIBUTING.md
• Closes fix(engine): local-scorer's totalTokenScore bypasses the TEST_FILE_CONTRIBUTION_WEIGHT discount when fed back into the score preview #8875

Validation

• git diff --check
• vitest run test/unit/local-scorer.test.ts (7 pass)
• vitest run test/unit/mcp-run-local-scorer.test.ts (2 pass)
• Engine/scorer-only; unrelated UI/workers checks not exercised.

Safety

• No secrets / private scores exposed
• No UI changes (N/A evidence)

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[loopover-orb]

This repository reviews pull requests one-shot: the PR must be correct as originally opened. Pushing an additional commit closes it automatically instead of restarting review — open a fresh pull request with every fix included.

[codecov]

❌ 3 Tests Failed:

Tests completed	Failed	Passed	Skipped
21941	3	21938	21

View the top 3 failed test(s) by shortest run time

test/unit/mcp-cli-plan-scorer-tools.test.ts > loopover-mcp plan-DAG + local-scorer + predict-gate tools (#6150) — pure tools > loopover_run_local_scorer computes source/test/non-code token scores from changed-file metadata

Stack Traces | 0.611s run time

AssertionError: expected 17.4 to be 25 // Object.is equality

- Expected
+ Received

- 25
+ 17.4

 ❯ test/unit/mcp-cli-plan-scorer-tools.test.ts:67:34

test/unit/miner-worktree-allocator-collisions.test.ts > loopover-miner worktree allocator collisions (#4298) > returns distinct worktree paths when multiple processes acquire simultaneously

Stack Traces | 0.853s run time

AssertionError: expected 2 to be 5 // Object.is equality

- Expected
+ Received

- 5
+ 2

 ❯ test/unit/miner-worktree-allocator-collisions.test.ts:148:41

test/unit/miner-worktree-allocator-collisions.test.ts > loopover-miner worktree allocator collisions (#4298) > rejects excess simultaneous cross-process acquire calls at the concurrency cap

Stack Traces | 0.926s run time

AssertionError: expected [ …(3) ] to have a length of 2 but got 3

- Expected
+ Received

- 2
+ 3

 ❯ test/unit/miner-worktree-allocator-collisions.test.ts:162:23

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}