fix: replay a session with a Write on the experimental ACP v2 surface (#1238)

Fixes a bug in #1237 on the opt-in ACP v2 surface. Nothing changes on
ACP v1.

## Problem

A v2 `session/resume` with `replayFrom` failed with an internal error on
any session that contains a Write:
- A replay has no approval preview and no PostToolUse hook, so the Write
reporter builds a diff from the input for every client on replay.
- That's a v1 diff. It even claims a creation, since the input doesn't
say whether the file existed.
- The v2 surface rejects v1 diffs (#1232), so the replay threw and the
resume failed.

`session-load-replay` has an Edit but no Write, so the scenario suite
didn't catch it.

## Change

A v2 client gets no diff from a Write's input, live or replayed: a v2
diff names the operation (`add` or `modify`), which the input doesn't
tell. Replayed Writes and Edits show no diff on v2 yet; building them
needs the tool's structured result, which the SDK's transcript reader
leaves out (see the follow-up discussion).

## Verification

A new test replays a transcript with a Write through a v2
`session/resume`: it failed with "Internal error" before the change and
passes after. The v1 golden files are unchanged. Build, lint and all
2386 tests pass.

src/tests/acp-v2.test.ts

@@ -740,6 +740,49 @@ describe("ACP v2 prompts", () => {
     ]);
   });
 
+  it("replays a Write without the v1 diff that v2 cannot take", async () => {
+    const client = v2Client();
+    await client.app.connectWith(connectRouter(), async (agent) => {
+      await initializeV2(agent);
+      const { sessionId } = await agent.request(v2.methods.agent.session.new, { cwd });
+      const file = path.join(cwd, "a.ts");
+      vi.mocked(getSessionMessages).mockResolvedValueOnce([
+        transcriptEntry("assistant", randomUUID(), {
+          id: "msg_write",
+          role: "assistant",
+          content: [
+            {
+              type: "tool_use",
+              id: "toolu_w",
+              name: "Write",
+              input: { file_path: file, content: "x\n" },
+            },
+          ],
+        }),
+        transcriptEntry("user", randomUUID(), {
+          role: "user",
+          content: [
+            {
+              type: "tool_result",
+              tool_use_id: "toolu_w",
+              content: `File created successfully at: ${file}`,
+            },
+          ],
+        }),
+      ] as unknown as Awaited<ReturnType<typeof getSessionMessages>>);
+      await agent.request(v2.methods.agent.session.resume, {
+        sessionId,
+        cwd,
+        replayFrom: { type: "start" },
+      });
+      await client.authUpdate(1);
+    });
+
+    expect(client.updates("tool_call_update")).toContainEqual(
+      expect.objectContaining({ toolCallId: "toolu_w", title: "Write a.ts", content: [] }),
+    );
+  });
+
   it("replays the history before it answers session/resume, clearing each message first", async () => {
     const echoed = scriptTurns(
       async function* (options) {

src/tool-calls/reporters/file-edit.ts

@@ -37,8 +37,10 @@ export class WriteReporter implements ToolReporter {
       // whether the file exists, so the live tool call shows no diff: a diff
       // without the old text would claim a creation. The adapter reads the
       // file only for the preview. A replay has no preview and no hook, so it
-      // keeps the diff.
-      if (!(capabilities.diffPatch || capabilities.v2) || replay) {
+      // keeps the diff, except for a v2 client: a v2 diff names the operation,
+      // which the input does not tell.
+      const inputDiff = capabilities.v2 ? false : !capabilities.diffPatch || replay;
+      if (inputDiff) {
         facts.change = [
           {
             type: "diff",