[Feature][Quant] Auto-detect quantization format from model files (#6645)

## Summary

- Add automatic quantization format detection, eliminating the need to
manually specify `--quantization` when serving quantized models.
- The detection inspects only lightweight JSON files
(`quant_model_description.json` and `config.json`) at engine
initialization time, with no `.safetensors` reads.
- User-explicit `--quantization` flags are always respected;
auto-detection only applies when the flag is omitted.

## Details

**Detection priority:**
1. `quant_model_description.json` exists → `quantization="ascend"`
(ModelSlim)
2. `config.json` contains `"quant_method": "compressed-tensors"` →
`quantization="compressed-tensors"` (LLM-Compressor)
3. Neither → default float behavior

**Technical approach:**
Hooked into `NPUPlatform.check_and_update_config()` to run detection
after `VllmConfig.__post_init__`. Since `quant_config` is already `None`
at that point, we explicitly recreate it via
`VllmConfig._get_quantization_config()` to trigger the full quantization
initialization pipeline.

## Files Changed

| File | Description |
|------|-------------|
| `vllm_ascend/quantization/utils.py` | Added
`detect_quantization_method()` and `maybe_auto_detect_quantization()` |
| `vllm_ascend/platform.py` | Integrated auto-detection in
`check_and_update_config()` |
| `vllm_ascend/quantization/modelslim_config.py` | Improved error
handling for weight loading |
- vLLM version: v0.15.0
- vLLM main:
d7e17aaacd

---------

Signed-off-by: SlightwindSec <slightwindsec@gmail.com>

tests/e2e/singlecard/test_quantization.py

@@ -49,6 +49,43 @@ def test_qwen3_w8a8_quant():
         name_1="vllm_quant_w8a8_outputs",
     )
 
+# fmt: off
+def test_qwen3_w8a8_quant_auto_detect():
+    """Test that ModelSlim quantization is auto-detected without --quantization.
+
+    Uses the same W8A8 model as test_qwen3_w8a8_quant but omits the
+    quantization parameter, verifying that the auto-detection in
+    maybe_auto_detect_quantization() picks up quant_model_description.json
+    and produces identical results.
+    """
+    max_tokens = 5
+    example_prompts = [
+        "vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs."
+    ]
+    vllm_target_outputs = [([
+        85, 4086, 44, 374, 264, 1550, 42747, 628, 323, 4938, 72816, 44378, 323,
+        13480, 4712, 369, 444, 10994, 82, 13, 1084, 374, 6188, 311, 387
+    ], 'vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. It is designed to be'
+                            )]
+# fmt: on
+
+    with VllmRunner(
+            "vllm-ascend/Qwen3-0.6B-W8A8",
+            max_model_len=8192,
+            gpu_memory_utilization=0.7,
+            cudagraph_capture_sizes=[1, 2, 4, 8],
+    ) as vllm_model:
+        vllm_quant_auto_detect_outputs = vllm_model.generate_greedy(
+            example_prompts, max_tokens)
+
+    check_outputs_equal(
+        outputs_0_lst=vllm_target_outputs,
+        outputs_1_lst=vllm_quant_auto_detect_outputs,
+        name_0="vllm_target_outputs",
+        name_1="vllm_quant_auto_detect_outputs",
+    )
+
+
 # fmt: off
 def test_qwen3_dense_w8a16():
     max_tokens = 5