docs(paged_attention): explain why seq_lens_host / block_table_host exist

The rationale (CANN CPU-tensor contract + NPUGraph capturability) was
only documented in the Ascend ATB kernel header.  Surface it on the base
class where the API contract lives, so any future backend implementor
understands why the optional host tensors are part of the signature.

Author: zhangyue

src/base/paged_attention.h

@@ -32,6 +32,19 @@ namespace infini::ops {
 //
 // Output layout:
 //   output      : [batch, num_heads, head_size]
+//
+// Optional host tensors: `seq_lens_host` and `block_table_host` are CPU
+// mirrors of `seq_lens` and `block_table`.  They exist because CANN's
+// paged-attention APIs mandate CPU-resident metadata — aclnn declares
+// `qSeqLens` as a CPU tensor in its signature, and ATB
+// `PagedAttentionParam` reads `aclIntArray*` parameters from the
+// `hostData` field at `aclnnRunner::Setup()` time.  Without caller-
+// provided host tensors, the kernel must synchronously D2H-copy both
+// each call, which (a) blocks the stream and (b) prevents NPUGraph
+// capture (sync copies are not capturable).  When the caller already
+// has CPU-pinned copies (e.g. vLLM's `optimistic_seq_lens_cpu` and
+// `BlockTable.get_cpu_tensor()`), passing them through lets the kernel
+// skip both D2H copies and be captured into a full NPUGraph.
 class PagedAttention : public Operator<PagedAttention> {
  public:
   PagedAttention(const Tensor query, const Tensor key_cache,