tensor: permuting axpy_to initializes an empty target

Tensor::axpy_to(arg, factor, perm) went straight to inplace_binary,
which asserts the target is non-empty. The first contribution into an
unallocated tensor -- e.g. a tensor-of-tensors contraction result inner
cell, when the inner op carries a permutation -- therefore aborted at
kernels.h:395. Mirror the non-permuting axpy_to overload: when the
target is empty, initialize it to factor * (perm ^ arg).

Fixes the einsum_manual/different_nested_ranks "ik;mn,j->ijk;nm" case
(mixed-nested-rank outer product with an inner-mode permutation) and
drops its TODO(tot-einsum-empty-result) breadcrumb.

Author: evaleev

src/TiledArray/tensor/tensor.h

@@ -2243,6 +2243,15 @@ class Tensor {
       return *this;
     } else {
       auto permuted = right.permute(perm);
+      if (empty()) {
+        // first contribution into an unallocated target (e.g. a contraction
+        // result inner cell): initialize to factor * (perm ^ arg) rather
+        // than asserting non-empty in inplace_binary -- mirrors the
+        // non-permuting axpy_to overload above.
+        *this = detail::clone_or_cast<Tensor>(permuted);
+        this->scale_to(factor);
+        return *this;
+      }
       return inplace_binary(
           permuted, [factor](auto& MADNESS_RESTRICT l, const auto& r) {
             using L = std::remove_reference_t<decltype(l)>;

tests/einsum.cpp

@@ -284,11 +284,6 @@ BOOST_AUTO_TEST_CASE(equal_nested_ranks) {
                                   {3, 2}));
 }
 
-// TODO(tot-einsum-empty-result): this case intermittently aborts at
-// tensor/kernels.h:395 ("!empty(result, tensors...)" / "0 of N tiles set").
-// Pre-existing on feature/arena_tensor HEAD -- reproduces with the arena/ToT
-// work stashed, and crashes deterministically when einsum_manual is run as a
-// standalone --run_test subset. Root cause not yet diagnosed.
 BOOST_AUTO_TEST_CASE(different_nested_ranks) {
   using ArrayT = TA::DistArray<TA::Tensor<int>>;
   using ArrayToT = TA::DistArray<TA::Tensor<TA::Tensor<int>>>;