§9 acceptance: grep-zero sweep — every internal x.ai/grok identifier renamed

The PRD's first acceptance gate now holds: grep -RinE '\bx\.ai\b|grok'
crates/ --include='*.rs' → 0 matches (exempt: NOTICE and third-party
license archives, README provenance, and the required 'Based on Grok
Build Open Source' attribution, now sourced from version_attribution.txt).

Wire-visible renames (both sides in this repo, changed in lockstep):
- Auth method id 'grok.com' → 'kimi-code' (AuthMethodKind::KimiCode).
- Every x.ai/* and _x.ai/* ACP ext method and meta key → kigi/* /
  _kigi/* (~200 names; grokShell → kigiShell). Session-file replay keeps
  a read-side alias for the legacy '_x.ai/session/update' method so
  existing updates.jsonl histories load; writes emit only the new name
  (both directions test-pinned).
- Agent types grok-build* → kigi* with a documented legacy-prefix alias
  at resolution time so persisted sessions keep resolving.
- ToolNamespace/BuiltinAgentName GrokBuild* → Kigi* (wire snake_case
  kigi/kigi_concise/kigi_hashline; schema regenerated); grok_build
  implementation dirs renamed to kigi*.
- x-grok-* headers → x-kigi-*, __GROK_* sentinels → __KIGI_*, themes
  grokday/groknight → kigiday/kiginight (old persisted values fall back
  to the default theme), web_fetch allowlist xAI hosts → kimi.com +
  moonshot platforms, changelog CDN → this repo, grok-build changelog
  archives deleted.
- BYOK default endpoint removed: [endpoints] api_base_url is now truly
  optional with NO default — consumers fail fast with the flag name when
  unset (no silent x.ai egress). Mock harnesses inject it explicitly.
- System-prompt identity fixed: 'released by xAI' → 'an unofficial
  community CLI for Kimi' (template + regenerated encrypted form).

Also repaired pre-existing grok-era test debt found by the sweep: the
stale trace_classify default-model pin, the grok-pager UA label test,
pty-harness stale-binary reuse and non-hermetic moonshot routing (a PTY
test could previously reach the real api.moonshot.cn), and the outdated
oauth fixture scope key.

Gates: §9 grep 0; fmt clean; workspace check/clippy 0/0 (-D warnings);
FULL cargo test --workspace: 234 suites, 21,961 passed, 0 failed;
deny advisories ok.

crates/codegen/kigi-tui/src/app/dispatch/turn.rs

@@ -258,7 +258,7 @@ pub(super) fn do_cancel_turn(app: &mut AppView, cancel_subagents: bool) -> Vec<E
 
     // Server-authoritative queue: the agent owns the drain. On an interactive
     // cancel we only tear down the running turn and let the agent promote the
-    // FRONT queued prompt as the next turn — its `x.ai/queue/changed`
+    // FRONT queued prompt as the next turn — its `kigi/queue/changed`
     // rebroadcast (carrying `running_prompt_id`) is the source of truth, and the
     // pager adopts it via `handle_queue_changed` / `apply_turn_start_shim`. We
     // do NOT pull any queued prompt back into the input or predict the new queue
@@ -277,7 +277,7 @@ pub(super) fn do_cancel_turn(app: &mut AppView, cancel_subagents: bool) -> Vec<E
     }]
 }
 
-/// Grace window between a driver-side `x.ai/session/prompt_complete`
+/// Grace window between a driver-side `kigi/session/prompt_complete`
 /// broadcast and that turn's `session/prompt` RPC response, after which
 /// [`reconcile_overdue_turn_ends`] finishes the turn from the broadcast. The
 /// healthy-path gap is milliseconds (the shell emits the broadcast just
@@ -285,7 +285,7 @@ pub(super) fn do_cancel_turn(app: &mut AppView, cancel_subagents: bool) -> Vec<E
 /// genuinely lost, not merely slow.
 pub(crate) const TURN_END_RECONCILE_GRACE: std::time::Duration = std::time::Duration::from_secs(2);
 
-/// Finish turns whose end was announced by `x.ai/session/prompt_complete`
+/// Finish turns whose end was announced by `kigi/session/prompt_complete`
 /// but whose `session/prompt` RPC response never arrived.
 ///
 /// The RPC response is the driver's only turn-state exit, and it can be lost