add PyStringMethods::encode_utf8

[davidhewitt]

This PR adds an encode_utf8() method to explicitly convert Bound<PyString> to Bound<PyBytes>, equivalent to the Python s.encode('utf-8').

This turns out to be useful in #3708 and I felt like it made sense for this to be a public API.

[davidhewitt]

I chose .encode_utf8() because I think in principle we could have .encode("some_encoding"), but I would prefer to punt on that more complete option as a future addition.

[codspeed-hq]

CodSpeed Performance Report

Merging #3801 will improve performances by 20.41%

_{Comparing davidhewitt:encode-utf8 (662eecf) with main (ecb4ecb)}

Summary

⚡ 3 improvements
✅ 76 untouched benchmarks

Benchmarks breakdown

	Benchmark	main	davidhewitt:encode-utf8	Change
⚡	mapping_from_dict	327.8 ns	272.2 ns	+20.41%
⚡	sequence_from_list	355.6 ns	300 ns	+18.52%
⚡	not_a_list_via_downcast	242.8 ns	215 ns	+12.92%

[Icxolu]

This looks good to me. I agree with you that encode is better used for a more generic encoding method. Also encode_utf8 makes it very clear which encoding is used.