attribute: add SLICE type support

[pellared]

Fixes #7934

$ go test -run=^$ -bench=BenchmarkSlice
goos: linux
goarch: amd64
pkg: go.opentelemetry.io/otel/attribute
cpu: 13th Gen Intel(R) Core(TM) i7-13800H
BenchmarkSlice/Len3/Value-20                    25297926                52.56 ns/op          144 B/op          1 allocs/op
BenchmarkSlice/Len3/KeyValue-20                 21315132                55.97 ns/op          144 B/op          1 allocs/op
BenchmarkSlice/Len3/AsSlice-20                  24214248                50.03 ns/op          144 B/op          1 allocs/op
BenchmarkSlice/Len3/String-20                   14148270                86.48 ns/op           48 B/op          1 allocs/op
BenchmarkSlice/Len3/Emit-20                     13605388                85.18 ns/op           48 B/op          1 allocs/op
BenchmarkSlice/Len5Nested/Value-20              16086171                71.30 ns/op          240 B/op          1 allocs/op
BenchmarkSlice/Len5Nested/KeyValue-20           15547844                75.81 ns/op          240 B/op          1 allocs/op
BenchmarkSlice/Len5Nested/AsSlice-20            17806996                66.16 ns/op          240 B/op          1 allocs/op
BenchmarkSlice/Len5Nested/String-20              7409064               165.2 ns/op            64 B/op          1 allocs/op
BenchmarkSlice/Len5Nested/Emit-20                7666302               161.0 ns/op            64 B/op          1 allocs/op
PASS
ok      go.opentelemetry.io/otel/attribute      12.980s

[codecov]

Codecov Report

❌ Patch coverage is 95.82173% with 15 lines in your changes missing coverage. Please review.
✅ Project coverage is 82.5%. Comparing base (56cd7ab) to head (22f4b51).

Files with missing lines	Patch %	Lines
trace/auto.go	0.0%	6 Missing ⚠️
attribute/value.go	98.7%	3 Missing and 1 partial ⚠️
...dk/metric/metricdata/metricdatatest/comparisons.go	0.0%	3 Missing ⚠️
attribute/hash.go	89.4%	2 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##            main   #8166     +/-   ##
=======================================
+ Coverage   82.3%   82.5%   +0.1%     
=======================================
  Files        309     309             
  Lines      24238   24513    +275     
=======================================
+ Hits       19963   20227    +264     
- Misses      3898    3909     +11     
  Partials     377     377             

Files with missing lines	Coverage Δ
attribute/key.go	100.0% <100.0%> (ø)
attribute/kv.go	100.0% <100.0%> (ø)
attribute/type_string.go	60.0% <ø> (ø)
log/keyvalue.go	97.1% <100.0%> (+<0.1%)	⬆️
attribute/hash.go	92.9% <89.4%> (+0.9%)	⬆️
...dk/metric/metricdata/metricdatatest/comparisons.go	96.9% <0.0%> (-0.6%)	⬇️
attribute/value.go	97.6% <98.7%> (+0.4%)	⬆️
trace/auto.go	95.6% <0.0%> (-1.7%)	⬇️

... and 3 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[dashpole]

Would it be more ergonomic for this to be variadic?

// variadic
myKey.Slice(IntValue(1), IntValue(2), IntValue(3))
// vs slice
myKey.Slice([]Value(IntValue(1), IntValue(2), IntValue(3)))

Not sure if we would want to change func Slice(k string, v []Value) KeyValue to func Slice(k string, v ...Value) KeyValue` as well though... so maybe consistency is more important?

[MrAlias]

+1 to variadic

[dashpole]

Do we have benchmarks for hashing for slices?