optimize: optimize serialization/deserialization of protocol codec

[slievrly]

• I have registered the PR changes.

Ⅰ. Describe what this PR did

optimize serialization/deserialization of protocol codec

Ⅱ. Does this pull request fix one issue?

fixes #6267

Ⅲ. Why don't you add test cases (unit test/integration test)?

Ⅳ. Describe how to verify it

Ⅴ. Special notes for reviews

[funky-eyes]

LGTM
The meaning of this configuration item needs to be modified on the document side

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 66.66667% with 6 lines in your changes are missing coverage. Please review.

Project coverage is 52.31%. Comparing base (f7fb1f8) to head (4d8925f).
Report is 5 commits behind head on 2.x.

❗ Current head 4d8925f differs from pull request most recent head f89a41b. Consider uploading reports for the commit f89a41b to get more accurate results

Additional details and impacted files

@@             Coverage Diff              @@
##                2.x    #6357      +/-   ##
============================================
+ Coverage     50.41%   52.31%   +1.90%     
+ Complexity     5252     5217      -35     
============================================
  Files           941      920      -21     
  Lines         33279    31954    -1325     
  Branches       4035     3818     -217     
============================================
- Hits          16776    16716      -60     
+ Misses        14877    13592    -1285     
- Partials       1626     1646      +20     

Files	Coverage Δ
.../apache/seata/core/protocol/ProtocolConstants.java	100.00% <100.00%> (+100.00%)	⬆️
...che/seata/core/rpc/netty/v1/ProtocolV1Decoder.java	61.81% <25.00%> (+47.27%)	⬆️
...seata/core/serializer/SerializerServiceLoader.java	52.63% <76.92%> (+52.63%)	⬆️

... and 87 files with indirect coverage changes

[funky-eyes]

I suggest to make some common and safe serialization types as defaults in this pr, instead of just one seata type.