[Spec] Add DSpark: confidence-scheduled speculative decoding (#30261)

Co-authored-by: sglang-bot <232288953+sglang-bot@users.noreply.github.com>
Co-authored-by: Claude Code <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Liangsheng Yin <lsyincs@gmail.com>
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>

test/registered/unit/sampling/test_sampling_batch_info.py

@@ -30,6 +30,7 @@ def _make_info(batch_size=2, **overrides):
         top_ks=torch.full((batch_size,), TOP_K_ALL, dtype=torch.int32),
         min_ps=torch.zeros(batch_size),
         is_all_greedy=False,
+        is_any_greedy=False,
         need_top_p_sampling=False,
         need_top_k_sampling=False,
         need_min_p_sampling=False,