passagemath-bliss: Add version 10.8.13 - #4795

Split out of the passagemath/passagemath monorepo (a SageMath fork) at
pkgs/sagemath-bliss, tagged passagemath-10.8.13, same shape as
build-passagemath-objects.yml. Vendors bliss (LGPL-3.0-or-later), a
graph-automorphism C++ library pinned to scipopt/bliss commit eddc188
with an already-verified sha256 for its source zip (build/pkgs/bliss's
own checksums.ini pin).

The sdist is built by replaying build/pkgs/sagemath_bliss/spkg-src
(./bootstrap renders pyproject.toml.m4, then python -m build --sdist
--no-isolation --skip-dependency-check against the real upstream
checkout). CIBW_BEFORE_ALL_LINUX replays build/pkgs/bliss/spkg-install.in
(cmake -GNinja -DUSE_GMP=OFF): EPEL ships no riscv64 repo at all, so
bliss-devel is never an option and this always builds from the pinned
source, same as upstream's own spkg-configure.m4 fallback.

Neither passagemath-bliss nor passagemath-objects ships a LICENSE file
for its own GPL-2.0-or-later code (0001, same gap/fix as objects); bliss
itself is LGPL-3.0-or-later (COPYING + COPYING.LESSER, the standard FSF
pair) and is fetched at build time rather than vendored in the sdist, so
nothing carries its licence text either (0002, drops both as
LICENSE.bliss).

sagemath-repl/-graphs (the real test extras) are unported, so
CIBW_TEST_COMMAND calls canonical_form() on a minimal duck-typed graph
instead.

- sage_setup's recursive Cython-module discovery requires an
  all__sagemath_bliss.py marker in EVERY package directory it walks
  (sage_setup.find.find_python_sources / is_package_or_sage_namespace_package_dir),
  not just at the sage/ root; without one in sage/graphs/ the sdist's
  pyproject.toml never even got rendered in the manylinux leg's log, and
  the module would not have been found either way. Added it to the
  sparse-checkout (confirmed present at that exact path in the real
  published sdist on PyPI).
- ./bootstrap's own per-package pyproject.toml.m4 render runs through
  sage-package's xargs-parallelised, ~650-package sweep, which aborts a
  whole batch of 8 if any member fails; sage_conf's own bootstrap always
  fails on this sparse checkout (needs the top-level configure.ac), and
  landed in the same batch as sagemath_bliss, so pyproject.toml was never
  written and the wheel build failed with 'No module named sage_setup'.
  Render this one package's pyproject.toml.m4 directly instead of
  trusting that sweep.
- musllinux (Alpine) has no dnf/no /usr/lib64; dnf install failed
  outright with 'dnf: not found'. Detect apk vs dnf, use Alpine's
  package names, and don't hard-fail if ldconfig isn't present.