Enhanced benchmarking for file_input

[djaglowski]

This PR extends #165. It reworks the file input operator's benchmarking tests, and adds additional scenarios that we should keep an eye on as we refactor the operator.

[djaglowski]

> make bench
go test -benchmem -run=^$ -bench ^* ./...
...
BenchmarkFileInput/Single-8               468019              2152 ns/op             688 B/op         10 allocs/op
BenchmarkFileInput/Glob-8                 110001             11013 ns/op            2762 B/op         42 allocs/op
BenchmarkFileInput/MultiGlob-8            113391             10673 ns/op            2762 B/op         42 allocs/op
BenchmarkFileInput/MaxConcurrent-8         42062             23964 ns/op            2757 B/op         42 allocs/op
BenchmarkFileInput/FngrPrntLarge-8        668179              2124 ns/op             689 B/op         10 allocs/op
BenchmarkFileInput/FngrPrntSmall-8        658712              1890 ns/op             688 B/op         10 allocs/op
PASS
ok      github.com/open-telemetry/opentelemetry-log-collection/operator/builtin/input/file      48.597s

[jsirianni]

My bench results using a Ryzen 9 3950x (16 cores, 32 thread)

go test -benchmem -run=^$ -bench ^* ./...
...
goos: linux
goarch: amd64
pkg: github.com/open-telemetry/opentelemetry-log-collection/operator/builtin/input/file
cpu: AMD Ryzen 9 3950X 16-Core Processor            
BenchmarkFileInput/Single-32         	  801732	      1347 ns/op	     640 B/op	       7 allocs/op
BenchmarkFileInput/Glob-32           	  335818	      3400 ns/op	    2561 B/op	      30 allocs/op
BenchmarkFileInput/MultiGlob-32      	  333190	      3426 ns/op	    2561 B/op	      30 allocs/op
BenchmarkFileInput/MaxConcurrent-32  	  191235	      5513 ns/op	    2561 B/op	      30 allocs/op
BenchmarkFileInput/FngrPrntLarge-32  	  786253	      1362 ns/op	     640 B/op	       7 allocs/op
BenchmarkFileInput/FngrPrntSmall-32  	  770312	      1376 ns/op	     640 B/op	       7 allocs/op