[ENHANCEMENT]: Update build-time check since the trunk CCCL allows 1B/2B atomic_ref

[PointKernel]

Is your feature request related to a problem? Please describe.

https://godbolt.org/z/a78x4qrha

Originally posted by @sleeepyjack #549 (comment)

Now cuda::atomic_ref can work with 1-byte and 2-byte variables so we need to update the static asserts accordingly in cuco data structures, e.g.

cuCollections/include/cuco/static_map_ref.cuh

Lines 77 to 78 in a7f87ac

	static_assert(sizeof(T) == 4 or sizeof(T) == 8,
	"sizeof(mapped_type) must be either 4 bytes or 8 bytes.");

cuCollections/include/cuco/detail/open_addressing/open_addressing_ref_impl.cuh

Line 1239 in e7b5a38

using packed_type = cuda::std::conditional_t<sizeof(value_type) == 4, uint32_t, uint64_t>;

Describe the solution you'd like

Update build checks and unit tests

Describe alternatives you've considered

No response

Additional context

No response

[sleeepyjack]

Having the same build time checks all over the library isn't ideal. Maybe this is an opportunity to refactor this whole insertion logic into a separate struct slot_inserter that internally decides which insertion strategy to use and also checks if the input types can be handled at all. Let me draft something in Godbolt.

[sleeepyjack]

Something like this: https://godbolt.org/z/nvGGYaPWa

This would also solve #547