Encoder optimizations

[gz-c]

Fixes (some of) #1878

Changes:

• Replace dchk in Deserialize methods with error handling during deserialization
• Optimize ParseTag method to remove memory allocations (strings.Split)
• Optimize reflect.Slice deserialization
• Optimize reflect.Array deserialization for byte arrays
• Optimize reflect.Slice serialization for arrays of fixed size numeric types
• Optimize reflect.Array serialization for arrays of fixed size numeric types

Does this change need to mentioned in CHANGELOG.md?
No

There are probably some more optimizations in here, the biggest ones might require caching which has its own issues. At the very least, the results of reflecting the struct for datasizeWrite could be shared with encoder.value. Also, determining the buffer length for slices of structs could be optimized if we can detect that the struct is composed of entirely fixed size objects (no strings, no slices).

It might be that the cost of datasizeWrite is higher than the cost of using a buffer that grows as needed. This is probably true, except maybe for the simplest structs.

Better gains will be had from replacing reflection with code generation.

Benchmark results:

# go test -bench=. -run=XXX -benchmem=true  ./src/daemon

benchmark                                        old ns/op     new ns/op     delta
BenchmarkDeserializeRawGetPeersMessage-8         155           78.3          -49.48%
BenchmarkSerializeGetPeersMessage-8              139           140           +0.72%
BenchmarkDeserializeRawGivePeersMessage-8        2790          810           -70.97%
BenchmarkSerializeGivePeersMessage-8             2121          1360          -35.88%
BenchmarkDeserializeRawIntroductionMessage-8     1548          615           -60.27%
BenchmarkSerializeIntroductionMessage-8          2886          1237          -57.14%
BenchmarkDeserializeRawGiveBlocksMessage-8       235440        39867         -83.07%
BenchmarkSerializeGiveBlocksMessage-8            216667        52153         -75.93%
BenchmarkDeserializeRawAnnounceTxnsMessage-8     4443          662           -85.10%
BenchmarkSerializeAnnounceTxnsMessage-8          4104          766           -81.34%
BenchmarkDeserializeRawGiveTxnsMessage-8         67512         11008         -83.69%
BenchmarkSerializeGiveTxnsMessage-8              61087         14769         -75.82%

benchmark                                        old allocs     new allocs     delta
BenchmarkDeserializeRawGetPeersMessage-8         2              1              -50.00%
BenchmarkSerializeGetPeersMessage-8              2              2              +0.00%
BenchmarkDeserializeRawGivePeersMessage-8        41             11             -73.17%
BenchmarkSerializeGivePeersMessage-8             31             17             -45.16%
BenchmarkDeserializeRawIntroductionMessage-8     22             7              -68.18%
BenchmarkSerializeIntroductionMessage-8          21             13             -38.10%
BenchmarkDeserializeRawGiveBlocksMessage-8       1277           392            -69.30%
BenchmarkSerializeGiveBlocksMessage-8            907            557            -38.59%
BenchmarkDeserializeRawAnnounceTxnsMessage-8     17             8              -52.94%
BenchmarkSerializeAnnounceTxnsMessage-8          7              8              +14.29%
BenchmarkDeserializeRawGiveTxnsMessage-8         377            116            -69.23%
BenchmarkSerializeGiveTxnsMessage-8              259            161            -37.84%

benchmark                                        old bytes     new bytes     delta
BenchmarkDeserializeRawGetPeersMessage-8         16            8             -50.00%
BenchmarkSerializeGetPeersMessage-8              16            16            +0.00%
BenchmarkDeserializeRawGivePeersMessage-8        689           160           -76.78%
BenchmarkSerializeGivePeersMessage-8             384           160           -58.33%
BenchmarkDeserializeRawIntroductionMessage-8     384           112           -70.83%
BenchmarkSerializeIntroductionMessage-8          320           160           -50.00%
BenchmarkDeserializeRawGiveBlocksMessage-8       51146         16584         -67.58%
BenchmarkSerializeGiveBlocksMessage-8            15728         11760         -25.23%
BenchmarkDeserializeRawAnnounceTxnsMessage-8     1127          352           -68.77%
BenchmarkSerializeAnnounceTxnsMessage-8          176           240           +36.36%
BenchmarkDeserializeRawGiveTxnsMessage-8         15463         4856          -68.60%
BenchmarkSerializeGiveTxnsMessage-8              4496          3408          -24.20%

# go test -bench=. -run=XXX -benchmem=true ./src/cipher/encoder

benchmark                     old ns/op     new ns/op     delta
BenchmarkDeserializeRaw-8     3065          657           -78.56%
BenchmarkSerialize-8          1069          790           -26.10%

benchmark                     old allocs     new allocs     delta
BenchmarkDeserializeRaw-8     31             11             -64.52%
BenchmarkSerialize-8          13             7              -46.15%

benchmark                     old bytes     new bytes     delta
BenchmarkDeserializeRaw-8     883           216           -75.54%
BenchmarkSerialize-8          208           112           -46.15%