Expose ReadOnlyNetworkGraph::get_addresses to C by cloning result

[TheBlueMatt]

We cannot expose ReadOnlyNetworkGraph::get_addresses as is in C as
it returns a list of references to an enum, which the bindings
dont support. Instead, we simply clone the result so that it
doesn't contain references.

There's a few options for how to deal with this, one suggested here. We could:

• take this as-is,
• have a cfg-flag for building bindings which switches the method's return type and adds the clone,
• eat the cost and just always clone.

I admittedly don't actually feel super strongly for any of them, just have one here to demonstrate.

[codecov]

Codecov Report

Merging #1115 (d2b7f6c) into main (d665748) will increase coverage by 0.37%.
The diff coverage is 0.00%.

@@            Coverage Diff             @@
##             main    #1115      +/-   ##
==========================================
+ Coverage   90.67%   91.05%   +0.37%     
==========================================
  Files          66       66              
  Lines       34684    36648    +1964     
==========================================
+ Hits        31449    33369    +1920     
- Misses       3235     3279      +44     

Impacted Files	Coverage Δ
lightning/src/routing/network_graph.rs	92.28% <0.00%> (+1.02%)	⬆️
lightning/src/ln/functional_tests.rs	97.84% <0.00%> (+0.40%)	⬆️
lightning/src/ln/onion_utils.rs	95.33% <0.00%> (+0.41%)	⬆️
lightning/src/ln/onion_route_tests.rs	97.76% <0.00%> (+0.66%)	⬆️
lightning/src/ln/functional_test_utils.rs	97.03% <0.00%> (+1.94%)	⬆️
lightning/src/util/events.rs	34.15% <0.00%> (+2.09%)	⬆️
lightning/src/ln/channelmanager.rs	87.81% <0.00%> (+2.66%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d665748...d2b7f6c. Read the comment docs.

[jkczyz]

I'm somewhat partial to cloning. I don't see this used in the code. How do we expect it to be used by users?

[TheBlueMatt]

I'm somewhat partial to cloning

Yea, fair enough, I don't think its worth being annoying about performance here - users shouldn't really be querying this aggressively...

I don't see this used in the code. How do we expect it to be used by users?

The #[doc(hidden)] is entirely ignored by the C bindings generation, so the function is exposed as normal.

[TheBlueMatt]

Changed to always cloning the result and exposing the original function, I think its cleaner.