refactor: simplify decoder & handler, add tests

[abemedia]

• Split request parsing out of handler.
• Reduce duplicate code in decoder compile.
• Remove type specific decoders.
• Remove tag check.
• Improve decoder test coverage.

[codecov]

Codecov Report

Patch coverage: 81.48% and project coverage change: +7.16 🎉

Comparison is base (f0c21df) 73.74% compared to head (c1cf53e) 80.90%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #89      +/-   ##
==========================================
+ Coverage   73.74%   80.90%   +7.16%     
==========================================
  Files          23       19       -4     
  Lines         899      770     -129     
==========================================
- Hits          663      623      -40     
+ Misses        201      119      -82     
+ Partials       35       28       -7     

Impacted Files	Coverage Δ
decoder/decode.go	56.25% <43.75%> (+56.25%)	⬆️
decoder/adapters.go	69.38% <69.38%> (ø)
decoder/compile.go	83.33% <95.45%> (+5.02%)	⬆️
encoding/form/form.go	72.72% <100.00%> (ø)
handler.go	96.61% <100.00%> (-1.31%)	⬇️
request.go	100.00% <100.00%> (ø)

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[github-actions]

Benchmark Result

Benchmark diff with base branch

goos: linux
goarch: amd64
pkg: github.com/abemedia/go-don
cpu: Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
          │ bench-master.txt │           bench-new.txt            │
          │      sec/op      │   sec/op     vs base               │
Handler-2        470.0n ± 0%   463.2n ± 1%  -1.44% (p=0.001 n=10)

          │ bench-master.txt │         bench-new.txt          │
          │       B/op       │    B/op     vs base            │
Handler-2         72.00 ± 0%   72.00 ± 0%  ~ (p=1.000 n=10) ¹
¹ all samples are equal

          │ bench-master.txt │         bench-new.txt          │
          │    allocs/op     │ allocs/op   vs base            │
Handler-2         2.000 ± 0%   2.000 ± 0%  ~ (p=1.000 n=10) ¹
¹ all samples are equal

pkg: github.com/abemedia/go-don/decoder
                    │ bench-master.txt │           bench-new.txt            │
                    │      sec/op      │   sec/op     vs base               │
Decoder/Gorilla-2          19.03µ ± 1%   19.13µ ± 1%  +0.53% (p=0.008 n=10)
Decoder/DonCached-2        942.7n ± 2%   947.1n ± 1%  +0.46% (p=0.015 n=10)
Decoder/Don-2              987.1n ± 0%   994.2n ± 1%  +0.72% (p=0.002 n=10)
geomean                    2.606µ        2.621µ       +0.57%

                    │ bench-master.txt │             bench-new.txt             │
                    │       B/op       │     B/op      vs base                 │
Decoder/Gorilla-2         2.500Ki ± 0%   2.500Ki ± 0%       ~ (p=1.000 n=10) ¹
Decoder/DonCached-2         176.0 ± 0%     176.0 ± 0%       ~ (p=1.000 n=10) ¹
Decoder/Don-2               176.0 ± 0%     176.0 ± 0%       ~ (p=1.000 n=10) ¹
geomean                     429.6          429.6       +0.00%
¹ all samples are equal

                    │ bench-master.txt │            bench-new.txt            │
                    │    allocs/op     │ allocs/op   vs base                 │
Decoder/Gorilla-2           166.0 ± 0%   166.0 ± 0%       ~ (p=1.000 n=10) ¹
Decoder/DonCached-2         3.000 ± 0%   3.000 ± 0%       ~ (p=1.000 n=10) ¹
Decoder/Don-2               3.000 ± 0%   3.000 ± 0%       ~ (p=1.000 n=10) ¹
geomean                     11.43        11.43       +0.00%
¹ all samples are equal

Benchmark result

goos: linux
goarch: amd64
pkg: github.com/abemedia/go-don
cpu: Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
          │ bench-new.txt │
          │    sec/op     │
Handler-2     463.2n ± 1%

          │ bench-new.txt │
          │     B/op      │
Handler-2      72.00 ± 0%

          │ bench-new.txt │
          │   allocs/op   │
Handler-2      2.000 ± 0%

pkg: github.com/abemedia/go-don/decoder
                    │ bench-new.txt │
                    │    sec/op     │
Decoder/Gorilla-2       19.13µ ± 1%
Decoder/DonCached-2     947.1n ± 1%
Decoder/Don-2           994.2n ± 1%
geomean                 2.621µ

                    │ bench-new.txt │
                    │     B/op      │
Decoder/Gorilla-2      2.500Ki ± 0%
Decoder/DonCached-2      176.0 ± 0%
Decoder/Don-2            176.0 ± 0%
geomean                  429.6

                    │ bench-new.txt │
                    │   allocs/op   │
Decoder/Gorilla-2        166.0 ± 0%
Decoder/DonCached-2      3.000 ± 0%
Decoder/Don-2            3.000 ± 0%
geomean                  11.43