refactor:Direct species node

[rprospero]

The SpeciesNode has been modified to output an actual "Species*" instead of "std::shared_ptr". This more closely matches the behaviour of the configuration node. Furthermore, the actual species is contained in a hidden parameter of the node, so there is no vestigial option that might confuse the user. This hidden parameter functionality can be used on other node types as well in the future for information that must be serialised but is not necessarily user facing.

This closes #2280 and #2277

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 2.

Benchmark suite	Current: 0b525a4	Previous: 9c38280	Ratio
BM_Box_MinimumImage<CubicBox>	44.437278100022034 ns/iter	14.652445275756 ns/iter	3.03
BM_Box_MinimumVector<CubicBox>	32.07641271673076 ns/iter	10.591956632536805 ns/iter	3.03
BM_Box_MinimumImage<OrthorhombicBox>	44.72856806946292 ns/iter	17.434331951610655 ns/iter	2.57
BM_Box_MinimumVector<OrthorhombicBox>	29.62962506850058 ns/iter	10.664159302468638 ns/iter	2.78
BM_Box_MinimumImage<MonoclinicAlphaBox>	34.650182638775306 ns/iter	11.827506507171602 ns/iter	2.93
BM_Box_MinimumVector<MonoclinicAlphaBox>	22.06917170875414 ns/iter	7.974686808099608 ns/iter	2.77
BM_Box_MinimumImage<TriclinicBox>	36.40650323010368 ns/iter	16.697681181962114 ns/iter	2.18
BM_Box_MinimumVector<TriclinicBox>	27.53808544794917 ns/iter	8.926373529580362 ns/iter	3.09

This comment was automatically generated by workflow using github-action-benchmark.

CC: @disorderedmaterials/dissolve-devs

[trisyoungs]

This all looks good, but I have one question - why does the SpeciesModel act as (from what I see) a producer of SpeciesNodes? I guess it depends on how we are utilising SpeciesModel in the UI, but In my head it should only ever be receiving the pointer to a SpeciesNode (or even just a Species...) from which it should be setting / getting data. I'm probably misunderstanding the intent here, so I had to ask!

[trisyoungs]

I might suggest that "internal" is a better definition than "hidden":

Suggested change

	// Serialise any hidden content
	virtual void hiddenSerialise(SerialisedValue &target) const {}
	// Deserialise any hidden content
	virtual void hiddenDeserialise(const SerialisedValue &node) {}
	// Serialise any required internal content
	virtual void serialiseInternal(SerialisedValue &target) const {}
	// Deserialise any required internal content
	virtual void deserialiseInternal(const SerialisedValue &node) {}