Deprecate cudf::strings::slice_strings APIs that accept delimiters

[davidwendt]

Description

Deprecating cudf::strings::slice_strings functions:
https://docs.rapids.ai/api/libcudf/stable/group__strings__slice.html#gaf1504116d31b0ec4f119f1477bb87ee1

std::unique_ptr<column> slice_strings(
  strings_column_view const& strings,
  string_scalar const& delimiter,
  size_type count,
  rmm::mr::device_memory_resource* mr);

and
https://docs.rapids.ai/api/libcudf/stable/group__strings__slice.html#ga21f01493d15c18d67b66a94f20a24389

std::unique_ptr<column> slice_strings(
  strings_column_view const& strings,
  strings_column_view const& delimiter_strings,
  size_type count,
  rmm::mr::device_memory_resource* mr);

These are not being used by cuDF (Cython) or Spark (JNI) and are quite convoluted and difficult to maintain.
Marking these as deprecated in 23.06 to be removed in 23.08 if possible.
Due to the warning/error compile settings, the benchmarks and the gtests are removed for these functions.

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[davidwendt]

/merge

[revans2]

@davidwendt we would like to use these in Spark. Would it be possible to un-deprecate them?

[davidwendt]

I think you could use cudf::strings::find()/rfind() to get the position values of the delimiter and then pass those positions appropriately to the current cudf::strings::slice_strings() to achieve the same result.