P0.4: wire the logic verification gate (sx assert helper + snapshot runner)

Add a real logic-test gate so future pure-sx game logic fails the build on a
bad assertion:

- tests/test.sx: `expect(cond, msg)` assert helper — prints a greppable
  `FAIL <file>:<line>: <msg>` and exits non-zero via process.exit on failure.
- tools/run_tests.sh: snapshot runner mirroring sx/tests/run_examples.sh; runs
  each tests/<name>.sx and diffs stdout + exit code against tests/expected/.
  Exits 0 iff all tests pass.
- tests/arith.sx (+ expected snapshots): seed passing sanity test.
- README.md: document both halves of the gate — logic runner and the
  reproducible ios-sim build/launch sequence (with device discovery).

tools/run_tests.sh

@@ -0,0 +1,112 @@
+#!/bin/bash
+# m3te logic-test gate. Mirrors sx/tests/run_examples.sh: run each sx logic
+# test, diff its stdout + exit code against committed expectations. Exits 0 iff
+# ALL tests pass, NON-ZERO otherwise (so a broken assertion fails the build gate).
+#
+# Usage: bash tools/run_tests.sh [--update]
+#   --update: regenerate the expected .stdout/.exit from current output.
+#
+# Layout (a test is any tests/<name>.sx with a tests/expected/<name>.exit marker;
+# helper modules such as tests/test.sx have no marker, so they are not run):
+#   tests/<name>.sx
+#   tests/expected/<name>.stdout   # normalized stdout
+#   tests/expected/<name>.exit     # process exit code
+#
+# SX binary: override with the SX env var; defaults to the dev build.
+
+set -uo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+ROOT_DIR="$(cd "$SCRIPT_DIR/.." && pwd)"
+SX="${SX:-/Users/agra/projects/sx/zig-out/bin/sx}"
+TESTS_DIR="$ROOT_DIR/tests"
+EXPECTED_DIR="$TESTS_DIR/expected"
+TIMEOUT=30
+
+PASS=0
+FAIL=0
+SKIP=0
+TIMEOUT_COUNT=0
+UPDATE=0
+
+if [[ "${1:-}" == "--update" ]]; then
+    UPDATE=1
+fi
+
+if [[ ! -x "$SX" ]]; then
+    echo "sx binary not found/executable at: $SX (override with SX=/path/to/sx)" >&2
+    exit 2
+fi
+
+# Collapse the absolute test-file path back to its repo-relative form so the
+# `FAIL <file>:<line>` lines printed by tests/test.sx are checkout-location
+# independent. Applied identically to expected and actual, so it can only
+# reconcile location noise, never desync an otherwise-matching pair.
+normalize() {
+    sed -E -e "s#${ROOT_DIR}/##g"
+}
+
+TMP_ERR="$(mktemp)"
+trap 'rm -f "$TMP_ERR"' EXIT
+
+shopt -s nullglob
+markers=("$EXPECTED_DIR"/*.exit)
+if [[ ${#markers[@]} -eq 0 ]]; then
+    echo "no tests found under $EXPECTED_DIR (expected <name>.exit markers)" >&2
+    exit 2
+fi
+
+for exit_file in "${markers[@]}"; do
+    name=$(basename "$exit_file" .exit)
+    sx_file="$TESTS_DIR/${name}.sx"
+    out_file="$EXPECTED_DIR/${name}.stdout"
+
+    if [[ ! -f "$sx_file" ]]; then
+        printf "  %-40s SKIP (missing %s)\n" "$name" "tests/${name}.sx"
+        SKIP=$((SKIP + 1))
+        continue
+    fi
+
+    printf "  %-40s" "$name"
+    actual_out=$(timeout "$TIMEOUT" "$SX" run "$sx_file" 2>"$TMP_ERR" | normalize)
+    actual_exit=${PIPESTATUS[0]}
+
+    if [[ $actual_exit -eq 124 ]]; then
+        TIMEOUT_COUNT=$((TIMEOUT_COUNT + 1))
+        echo "TIMEOUT (>${TIMEOUT}s)"
+        continue
+    fi
+
+    if [[ $UPDATE -eq 1 ]]; then
+        printf '%s\n' "$actual_out" > "$out_file"
+        printf '%s\n' "$actual_exit" > "$exit_file"
+        echo "updated (exit=$actual_exit)"
+        continue
+    fi
+
+    expected_out=$(normalize < "$out_file" 2>/dev/null)
+    expected_exit=$(cat "$exit_file")
+
+    out_ok=true; exit_ok=true
+    [[ "$actual_out" == "$expected_out" ]] || out_ok=false
+    [[ "$actual_exit" == "$expected_exit" ]] || exit_ok=false
+
+    if $out_ok && $exit_ok; then
+        PASS=$((PASS + 1))
+        echo "ok"
+    else
+        FAIL=$((FAIL + 1))
+        echo "FAIL"
+        $out_ok || { echo "  --- stdout diff (expected vs actual) ---"; diff <(printf '%s\n' "$expected_out") <(printf '%s\n' "$actual_out") || true; }
+        $exit_ok || echo "  exit code: expected=$expected_exit actual=$actual_exit"
+        [[ -s "$TMP_ERR" ]] && { echo "  --- stderr ---"; cat "$TMP_ERR"; }
+    fi
+done
+
+if [[ $UPDATE -eq 1 ]]; then
+    echo "Updated all expected output files."
+    exit 0
+fi
+
+echo "$PASS passed, $FAIL failed, $SKIP skipped, $TIMEOUT_COUNT timed out"
+[[ $FAIL -eq 0 && $TIMEOUT_COUNT -eq 0 ]]