Fix smollm2 alias to point at SmolLM2-135M (v2) instead of SmolLM-135M (v1) (#18859)

The original SmolLM2 PR (#9354) started as v1 support, was renamed to
`smollm2` during review, but the repo ID and `rope_theta` were never
updated to v2 values. The two checkpoints are genuinely different models
(0/272 tensors match).

- `HUGGING_FACE_REPO_IDS["smollm2"]`: `HuggingFaceTB/SmolLM-135M` →
`HuggingFaceTB/SmolLM2-135M`
- `examples/models/smollm2/135M_config.json`: `rope_theta` `10000.0` →
`100000.0` (matches [SmolLM2-135M HF
config](https://huggingface.co/HuggingFaceTB/SmolLM2-135M/blob/main/config.json))

### Test plan

Data-only change (one string, one number). Verified values match the
upstream HuggingFace SmolLM2-135M config.

Author: Copilot

examples/models/llama/export_llama_lib.py

@@ -123,7 +123,7 @@
     "qwen2_5_1_5b": "Qwen/Qwen2.5-1.5B",
     "qwen2_5_coder_32b": "Qwen/Qwen2.5-Coder-32B-Instruct",
     "phi_4_mini": "microsoft/Phi-4-mini-instruct",
-    "smollm2": "HuggingFaceTB/SmolLM-135M",
+    "smollm2": "HuggingFaceTB/SmolLM2-135M",
     "qwen3_0_6b": "Qwen/Qwen3-0.6B",
     "qwen3_1_7b": "Qwen/Qwen3-1.7B",
     "qwen3_4b": "Qwen/Qwen3-4B",

examples/models/smollm2/135M_config.json

@@ -6,7 +6,7 @@
     "n_kv_heads": 3,
     "n_layers": 30,
     "norm_eps": 1e-05,
-    "rope_theta": 10000.0,
+    "rope_theta": 100000.0,
     "use_scaled_rope": false,
     "vocab_size": 49152,
     "use_hf_rope": false,