Clippy fixes

[chris-ha458]

This PR ensures that even cargo clippy -- -Wclippy::pedantic yields no warnings.

One issue : I'm not sure if k_most_common_ordered can actually panic?
It seems like logically heap won't be empty.
If it can, maybe documentation could be better so it is understood in relation to outputs.
If not, maybe we can change it to unwrap and either document that it doesn't panic or suppress the lint

[coriolinus]

Why specify HashMap::default but not N::default?

[chris-ha458]

The honest answer is that because only hashmap triggered the clippy.
I think the current lint only looks for std or core traits.

Taking a closer look into the lint, it would be idiomatic to also change zero to N:: default.
added a commit for that.

[coriolinus]

This cannot in fact panic, because the heap is never empty. There are early returns in the function for if k == 0 or if k >= self.map.len().

[chris-ha458]

changed docs to reflect that it does not in fact panic as implemented

[coriolinus]

Changing an expect to an unwrap is a bad idea: if somehow the precondition does ever fail, we'll get better documentation of what we expected to hold true.

[coriolinus]

Is the clippy logic looking for the existence of a "Panics" section in the docs if there is a .expect in the function? I'd prefer to simply suppress that lint. Adding a big, bold-face documentation item "Panics: the current implementation does not panic" is unhelpful and pointless.

[chris-ha458]

I suppressed the lint with a comment that reflects how it doesnt panic currently.

[coriolinus]

That suppresses the lint right now, but it introduces distracting and potentially confusing documentation. The default expectation of a function is that it does not panic, unless it has a "Panics" section in the documentation. Adding a "Panics" section which just says "this does not panic" does not help someone reading the documentation, but it has the potential to confuse them. Better to omit that section of the documentation and instead suppress the lint with an #[allow(clippy::wants_pointless_documentation)] (not the actual lint name) annotation.

[chris-ha458]

oops sorry i was referring to eca86c5 I forgot to actually push before commenting.

i can also remove the insource documentation as well if you feel it is spurious

[coriolinus]

This is good, thanks!