add mthreads-ml-py dependency and refine the code

[gingerXue]

Overview

this PR introduces mthreads-ml-py as an explicit dependency and refacts the related detection code accordingly.

• Please refer the mthreads-ml-py pypi url

What’s Changed

• Added mthreads-ml-py to pyproject.toml as the official dependency for accessing MThreads runtime information.
• Refactored the detector structure to make platform-specific logic clearer and more modular.

Future Work

• Extend the MThreads detector method to provide more comprehensive device information by refering to nvidia's implementations.

Test Result

$ gpustack-runtime detect
+-------------------------------------------------------------+
| MTHREADS Driver Version: 3.3.2-server  Runtime Version: N/A |
|-------------------------------------------------------------|
| GPU |    Name   |    Memory-Usage   | GPU-Util | Temp |  CC |
|-----|-----------|-------------------|----------|------|-----|
| 0   | MTT S5000 | 201MiB / 81920MiB | 0%       | 38C  | N/A |
| 1   | MTT S5000 | 5MiB / 81920MiB   | 0%       | 44C  | N/A |
| 2   | MTT S5000 | 5MiB / 81920MiB   | 0%       | 42C  | N/A |
| 3   | MTT S5000 | 5MiB / 81920MiB   | 0%       | 39C  | N/A |
| 4   | MTT S5000 | 5MiB / 81920MiB   | 0%       | 39C  | N/A |
| 5   | MTT S5000 | 5MiB / 81920MiB   | 0%       | 43C  | N/A |
| 6   | MTT S5000 | 5MiB / 81920MiB   | 0%       | 42C  | N/A |
| 7   | MTT S5000 | 5MiB / 81920MiB   | 0%       | 39C  | N/A |
+-------------------------------------------------------------+

[yeahdongcn]

@thxCode Please take a look. Thanks.

[thxCode]

LGTM

[thxCode]

@gingerXue please execute make lint on your laptop before committing.

[thxCode]

@gingerXue One more thing, can you also check the output of gpustack-runtime topo and gpustack-runtime topo --format json? Thanks.

[gingerXue]

$ gpustack-runtime topo
+---------------------------------------------------------------------------------------------------------------------------------+
| MTHREADS | Device 0 | Device 1 | Device 2 | Device 3 | Device 4 | Device 5 | Device 6 | Device 7 | CPU Affinity | NUMA Affinity |
|----------|----------|----------|----------|----------|----------|----------|----------|----------|--------------|---------------|
| Device 0 | X        | NODE     | NODE     | NODE     | SYS      | SYS      | SYS      | SYS      | 0-31,64-95   | 0             |
| Device 1 | NODE     | X        | NODE     | NODE     | SYS      | SYS      | SYS      | SYS      | 0-31,64-95   | 0             |
| Device 2 | NODE     | NODE     | X        | NODE     | SYS      | SYS      | SYS      | SYS      | 0-31,64-95   | 0             |
| Device 3 | NODE     | NODE     | NODE     | X        | SYS      | SYS      | SYS      | SYS      | 0-31,64-95   | 0             |
| Device 4 | SYS      | SYS      | SYS      | SYS      | X        | NODE     | NODE     | NODE     | 32-63,96-127 | 1             |
| Device 5 | SYS      | SYS      | SYS      | SYS      | NODE     | X        | NODE     | NODE     | 32-63,96-127 | 1             |
| Device 6 | SYS      | SYS      | SYS      | SYS      | NODE     | NODE     | X        | NODE     | 32-63,96-127 | 1             |
| Device 7 | SYS      | SYS      | SYS      | SYS      | NODE     | NODE     | NODE     | X        | 32-63,96-127 | 1             |
+---------------------------------------------------------------------------------------------------------------------------------+

Legend (from nearest to farthest):
  X    = Self
  LINK = Connection traversing with High-Speed Link (e.g., AMD XGMI, Ascend HCCS, NVIDIA NVLink)
  PIX  = Connection traversing at most a single PCIe bridge
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  NODE = Connection traversing PCIe and the interconnect between NUMA nodes
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  N/A  = Unknown or unavailable information

$ gpustack-runtime topo --format json
[
  {
    "manufacturer": "mthreads",
    "devices_distances": [
      [
        0,
        40,
        40,
        40,
        50,
        50,
        50,
        50
      ],
      [
        40,
        0,
        40,
        40,
        50,
        50,
        50,
        50
      ],
      [
        40,
        40,
        0,
        40,
        50,
        50,
        50,
        50
      ],
      [
        40,
        40,
        40,
        0,
        50,
        50,
        50,
        50
      ],
      [
        50,
        50,
        50,
        50,
        0,
        40,
        40,
        40
      ],
      [
        50,
        50,
        50,
        50,
        40,
        0,
        40,
        40
      ],
      [
        50,
        50,
        50,
        50,
        40,
        40,
        0,
        40
      ],
      [
        50,
        50,
        50,
        50,
        40,
        40,
        40,
        0
      ]
    ],
    "devices_cpu_affinities": [
      "0-31,64-95",
      "0-31,64-95",
      "0-31,64-95",
      "0-31,64-95",
      "32-63,96-127",
      "32-63,96-127",
      "32-63,96-127",
      "32-63,96-127"
    ],
    "devices_numa_affinities": [
      "0",
      "0",
      "0",
      "0",
      "1",
      "1",
      "1",
      "1"
    ]
  }
]

[thxCode]

LGTM