perf: Faster singleton SiblingSubgraph construction

[aborgna-q]

Wraps the topoConvexChecker used by SiblingSubgraph in a OnceCell so we can delay initialization until it's needed.

This lets us avoid traversing the graph (and computing a topological order) when creating sibling subgraphs with zero or one nodes.

This partially helps with #1652.

See benchmark results:

$ critcmp before after -f '.*subgraph'
group                      after                                  before
-----                      -----                                  ------
singleton_subgraph/10      1.00      3.0±0.08µs        ? ?/sec    2.71      8.1±0.18µs        ? ?/sec
singleton_subgraph/100     1.00      4.8±0.07µs        ? ?/sec    9.81     47.4±1.16µs        ? ?/sec
singleton_subgraph/1000    1.00     23.2±1.86µs        ? ?/sec    18.94  439.3±17.04µs        ? ?/sec
multinode_subgraph/10      1.01     17.9±0.25µs        ? ?/sec    1.00     17.7±0.29µs        ? ?/sec
multinode_subgraph/100     1.01    170.0±3.72µs        ? ?/sec    1.00    168.5±3.04µs        ? ?/sec
multinode_subgraph/1000    1.02      2.4±0.02ms        ? ?/sec    1.00      2.3±0.04ms        ? ?/sec

singleton_subgraph creates a single-node subgraph in a region with around k * 3 nodes.
multinode_subgraph creates a subgraph with 1/3rd of the nodes for the same region.

This PR is quite noisy since it's adding those two new benchmarks, and tidying up the files in the process.

drive-by: Add bench = false for all the targets in the workspace. Otherwise, the auto-added test harness threw errors when passing criterion flags to cargo bench.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 85.51%. Comparing base (935c61b) to head (1c993b4).
Report is 2 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1654   +/-   ##
=======================================
  Coverage   85.50%   85.51%           
=======================================
  Files         136      136           
  Lines       25252    25264   +12     
  Branches    22164    22176   +12     
=======================================
+ Hits        21592    21604   +12     
  Misses       2456     2456           
  Partials     1204     1204           

Flag	Coverage Δ
python	92.42% <ø> (ø)
rust	84.55% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[doug-q]

Thank you!

[hugrbot]

This PR contains breaking changes to the public Rust API.
Please deprecate the old API instead (if possible), or mark the PR with a ! to indicate a breaking change.

cargo-semver-checks summary

--- failure auto_trait_impl_removed: auto trait no longer implemented ---

Description:
A public type has stopped implementing one or more auto traits. This can break downstream code that depends on the traits being implemented.
      ref: https://doc.rust-lang.org/reference/special-types-and-traits.html#auto-traits
     impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.36.0/src/lints/auto_trait_impl_removed.ron

Failed in:
type TopoConvexChecker is no longer Sync, in /home/runner/work/hugr/hugr/PR_BRANCH/hugr-core/src/hugr/views/sibling_subgraph.rs:457
type TopoConvexChecker is no longer RefUnwindSafe, in /home/runner/work/hugr/hugr/PR_BRANCH/hugr-core/src/hugr/views/sibling_subgraph.rs:457

[aborgna-q]

Uh, looks like I introduced a breaking change -.-
(TopoConvexChecker is no longer Sync now that it uses an OnceCell internally...)

[doug-q]

Uh, looks like I introduced a breaking change -.- (TopoConvexChecker is no longer Sync now that it uses an OnceCell internally...)

Awk. I think we can put it in a Mutex to restore unbrokenness?