perf(native): record pwntools diagnostics once in their Evidence (#1404)

inspect_binary_layout and inspect_recorded_crash passed the decoded report
as both result and rawResult, so every Evidence record carried the whole
report twice (and an MCP response four times, counting its text copy).
The decoded report is the only representation, so raw_result is now null,
as for other single-representation operations.

The real pwntools verifiers asserted the duplicate; they now assert null.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

src/application/binaryDiagnostics/BinaryLayoutService.test.ts

@@ -5,6 +5,7 @@ import {
   BINARY_LAYOUT_TEST_PROVIDER,
 } from "../../../tests/fixtures/binaryDiagnostics/layout.js";
 import { AnalysisCapabilityUnavailableError } from "../../domain/analysisErrorCore.js";
+import { parseEvidence } from "../../domain/evidence.js";
 import { err, ok } from "../../domain/result.js";
 
 it.each([
@@ -28,6 +29,19 @@ it.each([
   },
 );
 
+it("records the decoded layout once in its Evidence", async () => {
+  const layout = binaryLayoutFixture();
+  const service = new BinaryLayoutService({
+    identity: BINARY_LAYOUT_TEST_PROVIDER,
+    inspect: () => Promise.resolve(ok(layout)),
+  });
+  const result = await service.inspect({ path: layout.artifact.path });
+  if (!result.ok) throw result.error;
+  const evidence = parseEvidence(result.value);
+  expect(evidence.normalized_result).toEqual(layout);
+  expect(evidence.raw_result).toBeNull();
+});
+
 it("preserves a provider's actionable unsupported reason", async () => {
   const failure = new AnalysisCapabilityUnavailableError(
     "fixture-layout",