Adds a C++ e2e case proving re-entrancy: from inside an `on_echo_fired`
handler the consumer issues another request to the library, carrying data
taken from the event, and gets a correct response back.
The handler runs on the FFI thread with the event-registry lock held, so the
test documents and exercises the only safe shape: an *async* request. A
synchronous call from the handler would self-deadlock (the FFI thread is busy
running the handler), and add/removeEventListener would deadlock on the
registry lock. The async request merely queues on the FFI channel and is
drained once the handler returns; its future is moved out and resolved on the
main thread. A one-shot guard avoids the echo->event->echo storm (echo
re-fires the event). Timeouts turn any deadlock regression into a failure
rather than a hang.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* protect against mem leak in case of failures sending requests to ffi thread
* better cleanup if failures in createFFIContext
* avoid dangling cstring in handleRes under ARC/ORC
* better resource cleanup in destroyFFIContext
* invoke onNotResponding if failure in destroyFFIContext
* correct seq copy in alloc
* make sure the lock is init before cleanUpResources
* better possible exception handling in processReq
* guard allocSharedSeq if given seq is empty
* enhance error handling in ffi_context
* add new tests and some corrections
