Summarize BindingSpace trie trace shape

[MesTTo]

Stack dependency: review after #85. This branch is based on the prerequisite head. Until the prerequisite PRs merge, GitHub may include their commits in the aggregate diff; the intended review increment is the new commit at this branch tip.

Summarizes BindingSpace trie trace shape.

The branch adds TrieJoinTrace::summarize, TrieJoinTraceSummary, and shape validation errors for inconsistent trace metadata. The summary aggregates relation indexes, indexed rows, trie nodes, trace steps, candidate bindings, domain-source counts, domain-value counts, intersection sizes, empty intersections, bound-prefix depth, participating relation fan-in, and cursor counters.

Validation:

• cargo +nightly-2026-06-01 test -p mork --lib binding_plan with RUSTFLAGS='-C target-cpu=native'
• cargo +nightly-2026-06-01 test -p mork --lib binding_space with RUSTFLAGS='-C target-cpu=native'
• cargo +nightly-2026-06-01 test -p mork --lib with RUSTFLAGS='-C target-cpu=native'
• rustfmt --check on kernel/src/binding_plan.rs and kernel/src/binding_space.rs
• jscpd on kernel/src/binding_plan.rs, kernel/src/binding_space.rs, kernel/src/expression_trie.rs, kernel/src/pattern_relations.rs, kernel/src/test_exprs.rs, and kernel/src/test_sidecar_queries.rs
• git diff --check
• normal diffstat matched git diff -w --stat

[MesTTo]

Closing with the rest of the binding_space line (same reason as #84): the trace summary aggregates the parked sidecar join's traversal, not the engine's. An average-path-length/summary statistic on the real traversal, under a flag, is the rebuild worth doing.