fix(scheduler-sim-decision-parity): give sim mode driver.c's round-boundary split (task 2.1)

Replaces sim_loop.c's monolithic xce_core_go() with the same
xce_core_begin_session() -> sched_account_jobs() -> xce_core_signal_go()
sequence driver.c already uses for normal-mode rounds, opening the same
interleaving window and calling the now-core-level accounting service
(task 1.1) at the same relative position. Error-handling for a failed
session start is preserved byte-for-byte (same cleanup, same rc source).
Both interactive and batch simulation runs go through this one call site.

This is bug-1613's wiring half: sim_loop.c now calls sched_account_jobs()
at all (it never did before). resInUse is still zero end-to-end after this
task alone, because job->accounts is never populated by any live path in
sim mode yet -- that's Recorded-Decision Reconciliation (tasks 2.2-2.4),
not yet implemented; a comment marks where it slots in.

Adds a call-count/session-window regression test proving the wiring itself
(RED-phase reproduced live via git stash + rebuild, matching bug-1613's
symptom exactly). 8 sibling white-box test files needed a link-satisfying
sched_account_jobs() stub, discovered via actual linker errors.

45/46 scheduler tests pass; the 1 failure is pre-existing bug-1560
(unrelated).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

src/scheduler/testsuite/meson.build

@@ -726,6 +726,32 @@ sched_sim_equivalence_test = executable(
     dependencies: [scheduler_test_deps],
     install: false)
 
+# Round-boundary ordering contract for the simulation driver
+# (scheduler-sim-decision-parity spec, task 2.1, Req 1.3/1.4, 5.1, 5.3): the
+# sim-mode counterpart to sched_driver_session_ordering_test above -- proves
+# sim_loop.c's sched_sim_driver_run() now calls the relocated
+# sched_account_jobs() (core.h, task 1.1) once per replayed round, strictly
+# inside that round's session-active window, mirroring driver.c's
+# xce_sched_driver_run_normal_mode_round() exactly. Before this task,
+# sim_loop.c called the older monolithic xce_core_go() with no interleaving
+# slot and never called sched_account_jobs() at all (bug-1613) -- every
+# worker's/shared resource's resInUse stayed permanently zero for the whole
+# `--sim`/`--interactive` run. Same combined white-box #include set as
+# sched_sim_equivalence_test (records two real normal-mode rounds via the
+# REAL normal-mode driver, then replays the same file through the REAL,
+# unmodified simulation driver) -- see the test's own header comment for why
+# a spy sched_account_jobs() (not the real one) is the right-strength proof
+# here, and why no cross-thread wbuf-rendezvous replica of
+# test_driver_session_ordering.c's own design is attempted (sim_loop.c's
+# picker thread is not independently substitutable from a test without
+# touching sim_picker.c, outside this task's boundary).
+sched_sim_session_ordering_test = executable(
+    'sched_sim_session_ordering_test',
+    ['test_sim_session_ordering.c'],
+    include_directories:[incdir],
+    dependencies: [scheduler_test_deps],
+    install: false)
+
 # Resource-contention replay fixture (scheduler-sim-decision-parity spec,
 # task 1.2, Req 2.1/2.2): existing fixtures in this directory never have
 # scarce capacity, so none can exercise or prove capacity-check correctness.
@@ -1073,6 +1099,9 @@ test('scheduler sim record output test',
 test('scheduler sim equivalence test',
      sched_sim_equivalence_test, suite: 'scheduler',
      env: ['ASAN_OPTIONS=detect_leaks=0'])
+test('scheduler sim session ordering test',
+     sched_sim_session_ordering_test, suite: 'scheduler',
+     env: ['ASAN_OPTIONS=detect_leaks=0'])
 test('scheduler sim resource contention test',
      sched_sim_resource_contention_test, suite: 'scheduler',
      env: ['ASAN_OPTIONS=detect_leaks=0'])