fix(model_runner): correct seqlen_k to chunk boundary in prepare_prefill

During chunked prefill, seqlen_k was set to len(seq) (the full sequence
length), causing the attention kernel to access uninitialized KV slots
for tokens not yet scheduled in the current chunk.

Fix: reorder so that end = start + seqlen_q is computed first, then
set seqlen_k = end — limiting attention to the current chunk boundary.

Fixes #212

nanovllm/engine/model_runner.py

@@ -139,8 +139,8 @@ class ModelRunner:
             seqlen = len(seq)
             start = min(seq.num_cached_tokens, seqlen - 1)
             seqlen_q = seq.num_scheduled_tokens
-            seqlen_k = seqlen
             end = start + seqlen_q
+            seqlen_k = end
             input_ids.extend(seq[start:end])
             positions.extend(range(start, end))
             cu_seqlens_q.append(cu_seqlens_q[-1] + seqlen_q)