test(fs-search): re-record the glob-sampling snapshot against the real API

The scenario previously carried an authored fixture; W4 of #1119 review
requires a live transcript. Recording surfaced two composition bugs that
are fixed here alongside it:
- provider ids: the app and the replay catalog both named the old
  'deepseek' provider, which no adapter registers; both now use
  'deepseek-official'
- the live config lacked persistenceCompression: none, so record-mode
  sessions were written zstd-compressed and could not be harvested
  (the snapshot twin already forced plaintext)

Recorded logs also need deterministic replay:
- packChunks: false in both configs — the eager-drain batch boundaries
  that split packed delta runs are timing-dependent, so a packed log of
  a long reasoning stream cannot replay-match its live record
- the fixture's request/header config and request/context are normalized
  to the replay-produced minimal shape (the live adapter logs model
  capabilities llm-replay has no data for), and tool-result path
  separators are canonicalized to '/' for the Linux golden

posixOnly is restored now that the fixture is recorded.

examples/acp-agent/tests/fs-search.cordis.snapshot.yml

@@ -3,7 +3,7 @@
   name: '@deepseek-ai/dsh-llm-replay'
   config:
     providers:
-      - id: deepseek
+      - id: deepseek-official
         name: DeepSeek
         models:
           - id: deepseek-v4-pro
@@ -17,10 +17,14 @@
 - id: acp-agent
   name: '@deepseek-ai/dsh-acp-demo'
   config:
-    provider: deepseek
+    provider: deepseek-official
     model: deepseek-v4-pro
     persistenceRoot: !!js process.env.DSH_SNAPSHOT_SESSIONS_ROOT ?? './.sessions'
     persistenceCompression: none
+    # Unpacked rows: the eager-drain batch boundaries that split packed delta
+    # runs are timing-dependent, so packed logs cannot replay-match a live
+    # record of a long reasoning stream.
+    packChunks: false
     workspaceContext: false
     skills:
       enabled: false