fix: QA remediation P1-A/P1-B/P1-C/P2 for v0.22.1 sync-mode

P1-A: wrap wp_schedule_single_event(GENERATE_ACTION) + fire_cron_now() in
on_orchestrate_tick() with 'if ( ! self::$sync_mode )' guard.
Mirrors the existing guard in on_generate_tick(). Prevents a background
wp-cron process spawning in parallel with the VPS --sync loop and
prematurely finalizing the run lock before Article_Worker completes.

P1-B: add two unit tests to GenerationCheckpointRunnerTest.php:
- test_orchestrate_tick_in_sync_mode_does_not_schedule_background_event:
  asserts sync_mode=true suppresses GENERATE_ACTION schedule + spawn_cron.
- test_generate_tick_async_mode_schedules_generate_action:
  asserts sync_mode=false (async path) still reschedules GENERATE_ACTION.

P1-C: create root CONTEXT.md with domain glossary (run, run_stages,
checkpoint, idea queue, Generation_Lock) and execution mode reference
(async wp-cron vs --sync VPS-driven).

P2: update ARCHITECTURE.md §25 Entry Points table with the
'wp prautoblogger generate --sync' VPS path row.

Agent-Session: prab-reliability-remediation-20260622

ARCHITECTURE.md

@@ -1117,7 +1117,8 @@ If the host kills any tick:
 | Caller | Path |
 |--------|------|
 | Board "New Article" button | AJAX `prautoblogger_generate_now` → `Generation_Status_Poller::on_ajax_generate_now()` → schedules `prautoblogger_manual_generation` → `Executor::on_manual_generation()` → `kick_off()` |
-| WP-CLI | `wp prautoblogger generate` → `WP_CLI_Commands::generate_command()` → `kick_off()` |
+| WP-CLI (async) | `wp prautoblogger generate` → `WP_CLI_Commands::generate_command()` → `kick_off()` |
+| WP-CLI (sync / VPS) | `wp prautoblogger generate --sync` → `WP_CLI_Commands::run_sync()` → `set_sync_mode(true)` → `on_orchestrate_tick()` → tick loop `on_generate_tick()` (×N) → `set_sync_mode(false)` |
 | Old cron hook (retained) | `prautoblogger_manual_generation` → `Executor::on_manual_generation()` → `kick_off()` |
 
 ### SSH Workaround Retirement