[FEA] libcudf APIs accepting `std::vector<T> const&` should instead accept a span

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

Consider, for example, `cudf::merge`

```c++
std::unique_ptr merge(
std::vector const& tables_to_merge,
std::vector const& key_cols,
std::vector const& column_order,
std::vector const& null_precedence = {},
cuda::stream_ref stream = cudf::get_default_stream(),
rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
```

Suppose that I have a perfectly decent array of values for the `key_cols` argument, perhaps a `{cudf::size_type *, std::size_t}` pair. I would have this if, for example, I am in a compiled language and trying to interface with the libcudf API: I will do so via a C ABI-shim.

The current libcudf APIs require me to materialise, by copy, a `std::vector` from my span-like object.

**Describe the solution you'd like**

A putative signature:

```c++
std::unique_ptr merge(
std::span tables_to_merge,
std::span key_cols,
std::span column_order,
std::span null_precedence = {},
cuda::stream_ref stream = cudf::get_default_stream(),
rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
```

That way, if I have a `std::vector` I'm good, and equally if I have a `T const *, size` pair I'm also good.

**Describe alternatives you've considered**

Just make a local copy.

Start with the libcudf API represented by cudf::merge and identify other APIs accepting std::vector const&. Review how std::span would affect callers, including the C ABI-shim use case, then verify that affected declarations and existing vector-based calls remain compatible.

Written by the indexing model from the issue text.