[Doc][Attention] Clarify dense GQA input-inferred contract

[superAngGao]

Summary

• document that Dense GQA derives B, Sq, Skv, H, Hkv, D, and input dtype from every forward call
• separate construction-time attention semantics from call-time tensor metadata
• define the full computation: FP8 dequantization, optional RoPE, scaled QK, softcap, causal/window masking, softmax, and PV reduction
• map every constructor parameter to its effect in the equations
• clarify that one Op instance may resolve and cache implementations for multiple input shapes

Scope

This draft does not change the constructor, forward signature, manifest, validation, dispatch, or kernels. It only clarifies the input-inferred contract established in #1975, following the documentation structure used by GemmFwdOp.

Validation

• python -m py_compile src/tileops/ops/attention/gqa.py
• python -m ruff check src/tileops/ops/attention/gqa.py
• PYTHONUTF8=1 python scripts/lint/op_docstrings_lint.py
• git diff --check