fix(riscv): take pending interrupt trap in wfi

[carlosqwqqwq]

Make wfi take the interrupt trap immediately when an enabled interrupt is
pending, per priv 1.12 norm:wfi_mepc_val. Currently Op::Wfi always returns
Ok(RiscVExit::Wfi) (stall hint) even when mip & mie is non-zero.

Validation

• Probe R-014 (asserting the defective stall behavior) now fails, i.e. the
  defect is fixed.
• New regression tests v058/v058b verify: no pending interrupt -> still
  returns Wfi; pending enabled interrupt -> traps.
• Full probe suite: 53 passed; the 8 failures are all expected
  (R-010..R-014 for the sret/wfi fixes, R-023 for the segment-ff decode fix,
  R-017 obsolete, V-012 already adopted upstream).
• cargo test --lib passes.

Closes #290

[19h]

Thanks for the report and patch. I independently checked WFI against the privileged architecture and implemented the valid part in #296: a locally enabled pending interrupt wakes WFI irrespective of the global enable. The rollup does not force a trap when global/delegation conditions make that interrupt non-actionable, because wake eligibility and trap eligibility are distinct. It includes your co-author credit and asks for your review. I’m closing this PR as superseded by #296.