Delay initialization of io.netty.handler.codec.compression.BrotliDecoder to runtime

[gradinac]

Motivation:
See #11427

Modification:
By delaying the initialization of io.netty.handler.codec.compression.BrotliDecoder to runtime, native-image will not try to eagerly initialize the class during the image build, avoiding the build failure described in the issue.

Fixes #11427

[normanmaurer]

@gradinac can you sign our icla: https://netty.io/s/icla ?

[gradinac]

@normanmaurer I've requested an approval to sign it - will sign it and let you know as soon as I get a green light on it

[normanmaurer]

@chrisvest any idea why the build would fail with graal ?

[gradinac]

Looking at surefire reports of the build, this seems to be a crash in the JVM:

# Created at 2021-06-30T12:21:57.124
Corrupted STDOUT by directly writing to native stream in forked JVM 1. Stream '# A fatal error has been detected by the Java Runtime Environment:'.
# Created at 2021-06-30T12:21:57.124
Corrupted STDOUT by directly writing to native stream in forked JVM 1. Stream '#  SIGSEGV (0xb) at pc=0x00007f3ac8a088f2, pid=8110, tid=8123'.

These failures usually produce a hs_err file that can be very useful for debugging, though I'm not sure if it's picked up by the gate - for this particular crash, the file is located at:

# Created at 2021-06-30T12:21:57.124
Corrupted STDOUT by directly writing to native stream in forked JVM 1. Stream '# /code/resolver/hs_err_pid8110.log'.
java.lang.IllegalArgumentException: Stream stdin corrupted. Expected comma after third character in command '# /code/resolver/hs_err_pid8110.log'.

This is most likely a bug and will probably not happen again if we re-run the job. One thing I've noticed is that the GraalVM version used in the gate is very old:

# Created at 2021-06-30T12:21:57.124
Corrupted STDOUT by directly writing to native stream in forked JVM 1. Stream '# JRE version: OpenJDK Runtime Environment GraalVM CE 20.1.0 (11.0.7+10) (build 11.0.7+10-jvmci-20.1-b02)'.

It may be worth bumping it to 21.1 (or wait a bit more for 21.2 to be released) - the cause of this crash might have already been addressed in the newest version

[gradinac]

@normanmaurer I have signed the ICLA

[normanmaurer]

@gradinac thanks a lot!

[gradinac]

@normanmaurer thank you very much for reviewing and merging this PR! :)