Fix scheduler firing benchmark: no inner loop, warm-up call

- Revert component.h change (no core changes for benchmarks)
- Remove -DWARN_IF_BLOCKING_OVER_MS from build flags (can't shadow constexpr)
- Drop inner loop — 5 heap pops + callbacks + pushes per call is well
  above CodSpeed's 60ns instrumentation overhead
- Add warm-up call before benchmark loop to trigger the blocking guard
  once and ramp the threshold
- interval=0 causes infinite loop, must use interval=1 with fake time

script/cpp_benchmark.py

@@ -34,10 +34,6 @@ PLATFORMIO_OPTIONS = {
         "-O2",  # optimize for speed (CodSpeed recommends RelWithDebInfo)
         "-g",  # debug symbols for profiling
         USE_TIME_TIMEZONE_FLAG,
-        # Disable WarnIfComponentBlockingGuard check — scheduler benchmarks
-        # use fake monotonic time ahead of real millis(), causing uint32_t
-        # underflow in the guard's (millis() - started_) calculation.
-        "-DWARN_IF_BLOCKING_OVER_MS=UINT32_MAX",
     ],
     # Use deep+ LDF mode to ensure PlatformIO detects the benchmark
     # library dependency from nested includes.