CANN: Add support for async operator submission

[hipudding]

Submit operators using asynchronous threads to improve performance.

Use the environment variable GGML_CANN_ASYNC_MODE to control whether
asynchronous submission is enabled. It is disabled by default.

Testing shows a 10%–20% performance improvement in scenarios with
small parameter sizes, especially in quantized models.

SYNC_MODE

llama_perf_sampler_print:    sampling time =      76.81 ms /   316 runs   (    0.24 ms per token,  4113.94 tokens per second)
llama_perf_context_print:        load time =    2880.65 ms
llama_perf_context_print: prompt eval time =      23.05 ms /    27 tokens (    0.85 ms per token,  1171.11 tokens per second)
llama_perf_context_print:        eval time =    6727.99 ms /   288 runs   (   23.36 ms per token,    42.81 tokens per second)
llama_perf_context_print:       total time =    7838.36 ms /   315 tokens

ASYNC_MODE

llama_perf_sampler_print:    sampling time =      51.17 ms /   220 runs   (    0.23 ms per token,  4299.73 tokens per second)
llama_perf_context_print:        load time =    2751.20 ms
llama_perf_context_print: prompt eval time =      17.26 ms /    27 tokens (    0.64 ms per token,  1563.95 tokens per second)
llama_perf_context_print:        eval time =    3037.53 ms /   192 runs   (   15.82 ms per token,    63.21 tokens per second)
llama_perf_context_print:       total time =    3343.86 ms /   219 tokens

[noemotiovon]

This PR improves NPU utilization through asynchronous dispatching — impressive work!