pipeline: use Message value to reduce GC pressure

[overvenus]

What problem does this PR solve?

Use Message value to reduce GC pressure

# master
goos: linux
goarch: amd64
pkg: github.com/pingcap/ticdc/pkg/pipeline
cpu: Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz
BenchmarkPipeline/BenchmarkPipeline/1_node(s)-40         	 1309424	       819.7 ns/op	      56 B/op	       2 allocs/op
BenchmarkPipeline/BenchmarkPipeline/2_node(s)-40         	  953646	      1214 ns/op	      80 B/op	       3 allocs/op
BenchmarkPipeline/BenchmarkPipeline/3_node(s)-40         	  739050	      1639 ns/op	     104 B/op	       4 allocs/op
BenchmarkPipeline/BenchmarkPipeline/4_node(s)-40         	  631210	      1953 ns/op	     128 B/op	       5 allocs/op
BenchmarkPipeline/BenchmarkPipeline/5_node(s)-40         	  535172	      2387 ns/op	     152 B/op	       6 allocs/op
BenchmarkPipeline/BenchmarkPipeline/6_node(s)-40         	  429428	      2819 ns/op	     176 B/op	       7 allocs/op
BenchmarkPipeline/BenchmarkPipeline/7_node(s)-40         	  404947	      3508 ns/op	     200 B/op	       8 allocs/op
BenchmarkPipeline/BenchmarkPipeline/8_node(s)-40         	  326889	      3714 ns/op	     224 B/op	       9 allocs/op
PASS
ok  	github.com/pingcap/ticdc/pkg/pipeline	11.001s

# this patch
goos: linux
goarch: amd64
pkg: github.com/pingcap/ticdc/pkg/pipeline
cpu: Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz
BenchmarkPipeline/BenchmarkPipeline/1_node(s)-40         	 1643746	       732.4 ns/op	      48 B/op	       1 allocs/op
BenchmarkPipeline/BenchmarkPipeline/2_node(s)-40         	  972543	      1195 ns/op	      96 B/op	       2 allocs/op
BenchmarkPipeline/BenchmarkPipeline/3_node(s)-40         	  777621	      1634 ns/op	     144 B/op	       3 allocs/op
BenchmarkPipeline/BenchmarkPipeline/4_node(s)-40         	  648489	      1982 ns/op	     192 B/op	       4 allocs/op
BenchmarkPipeline/BenchmarkPipeline/5_node(s)-40         	  507319	      2488 ns/op	     240 B/op	       5 allocs/op
BenchmarkPipeline/BenchmarkPipeline/6_node(s)-40         	  395040	      2822 ns/op	     288 B/op	       6 allocs/op
BenchmarkPipeline/BenchmarkPipeline/7_node(s)-40         	  376586	      3316 ns/op	     336 B/op	       7 allocs/op
BenchmarkPipeline/BenchmarkPipeline/8_node(s)-40         	  318111	      3858 ns/op	     384 B/op	       8 allocs/op
PASS
ok  	github.com/pingcap/ticdc/pkg/pipeline	10.624s

Left: master
Right: this patch

Cc #2212

Check List

Tests

• Unit test
• Integration test

Related changes

• Need to cherry-pick to the release branch

Release note

None

[ti-chi-bot]

[REVIEW NOTIFICATION]

This pull request has been approved by:

• amyangfei
• liuzix

To complete the pull request process, please ask the reviewers in the list to review by filling /cc @reviewer in the comment.
After your PR has acquired the required number of LGTMs, you can assign this pull request to the committer in the list by filling /assign @committer in the comment to help you merge this pull request.

The full list of commands accepted by this bot can be found here.

Reviewer can indicate their review by submitting an approval review.
Reviewer can cancel approval by submitting a request changes review.

[amyangfei]

Could we add a benchmark in the unit test to verify the optimization for pipeline memory allocation

[overvenus]

/run-all-tests

[overvenus]

/hold

[overvenus]

/unhold

[overvenus]

/run-kafka-tests
/run-integration-tests

[amyangfei]

/merge

[ti-chi-bot]

This pull request has been accepted and is ready to merge.

Commit hash: 22d835f

[ti-chi-bot]

In response to a cherrypick label: new pull request created: #2439.

[ti-chi-bot]

In response to a cherrypick label: new pull request created: #2440.

[ti-chi-bot]

In response to a cherrypick label: new pull request created: #2441.