fix(0x11): send ECUReset positive response before performing the reset

[w1ne]

Closes #76.

Bug

ISO 14229-1 (ECUReset) requires:

The ECUReset positive response message (if required) shall be sent before the reset is executed in the server(s).

The handler did the reverse — it called fn_reset() first, then built and sent the 0x51 response:

/* Process Reset */
if (ctx->config->fn_reset) {
    ctx->config->fn_reset(ctx, sub);   /* reboots the MCU here, never returns */
}
...
return uds_send_response(ctx, 2u);     /* unreachable on real hardware */

On a real ECU, fn_reset performs the actual reset (e.g. NVIC_SystemReset()) and never returns, so the positive response was never put on the wire and the tester saw no answer.

Fix

Reorder: emit the positive response (or honour suppressPosRsp), then call fn_reset(). The 0x51 reply is 2 bytes (single ISO-TP frame) so it is fully handed to transport synchronously before the reset is performed.

Test

Added test_ecu_reset_response_sent_before_reset: local transport and reset callbacks stamp a shared monotonic counter; the test asserts send_order < reset_order (and the 0x51 01 payload). Fails before the fix, passes after.

Verification

• New regression test: red before, green after
• Full suite: 100% tests passed, 0 tests failed out of 59
• clang-format 18 --dry-run --Werror (CI Docker image) clean across src include examples and the test file

Note

This is the response-ordering defect. The related stuck-suppressPosRsp leak (next service after a 0x11 0x81 getting silently suppressed) is fixed separately in #81 — it lives in uds_core.c dispatch, so the two changes touch different files and don't conflict.