feat: add tensor dumping functionality with shape information

- Add --dump-folder CLI argument to enable tensor dumping during network communication
- Implement binary dump format with tensor shape metadata (n_embed, n_tokens)
- Dump both send and receive tensors with unique filenames and counters
- Include proper parameter passing from CLI to llama_send_tensors/llama_recv_tensors functions

The dump format includes: element_type(1B) + n_embed(8B) + n_tokens(8B) + tensor_size(8B) + data

include/llama.h

@@ -378,6 +378,9 @@ extern "C" {
         // currently works only with CPU execution
         ggml_abort_callback abort_callback;
         void *              abort_callback_data;
+
+        // Tensor dumping path - if provided, network communication tensors will be dumped
+        const char * dump_folder;
     };
 
     // model quantization parameters