After JDK-8242181, `runtime/ErrorHandling/TestDwarf.java` fails in GHA:

```
...
[dwarf] ##### Find filename and line number for offset 0x00085ff1 in library /lib32/libc.so.6 #####
[dwarf] Failed to load DWARF file for library /lib32/libc.so.6 or find DWARF sections directly inside it.
#
# Compiler replay data is saved as:
# /home/runner/work/jdk/jdk/build/run-test-prebuilt/test-support/jtreg_test_hotspot_jtreg_tier1_runtime/scratch/replay_pid5024.log
#
# If you would like to submit a bug report, please visit:
# https://bugreport.java.com/bugreport/crash.jsp
#

hs_err_file: hs_err_pid5024.log
----------System.err:(16/999)----------
java.lang.RuntimeException: Must find library in "C [libc.so.6+0x85ff1]": expected true, was false
at jdk.test.lib.Asserts.fail(Asserts.java:594)
at jdk.test.lib.Asserts.assertTrue(Asserts.java:486)
at TestDwarf.checkNoSourceLine(TestDwarf.java:181)
at TestDwarf.runAndCheck(TestDwarf.java:153)
at TestDwarf.test(TestDwarf.java:101)
at TestDwarf.main(TestDwarf.java:91)
at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
at java.base/java.lang.reflect.Method.invoke(Method.java:578)
at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127)
at java.base/java.lang.Thread.run(Thread.java:1589)
```

We might want to consider making this test more resilient for these cases, but it seems beneficial to let current GHA hosts run it too, by providing the packages carrying the missing .so file.