RFR: 8316156: (ch) Channels.newOutputStream(ch).write(bigByteArray) allocates a lot of direct memory [v3]

[Brian Burkhalter]

> In `ChannelInputStream` and `ChannelOutputStream`, process small chunks of bytes in sequence to avoid running afoul of the `MaxDirectMemorySize`.

Brian Burkhalter has updated the pull request incrementally with one additional commit since the last revision:

  8316156: Convert test to JUnit 5 and separate read/write testing

-------------

Changes:
  - all: https://git.openjdk.org/jdk/pull/15733/files
  - new: https://git.openjdk.org/jdk/pull/15733/files/a9af89cf..c336780d

Webrevs:
 - full: https://webrevs.openjdk.org/?repo=jdk&pr=15733&range=02
 - incr: https://webrevs.openjdk.org/?repo=jdk&pr=15733&range=01-02

  Stats: 59 lines in 1 file changed: 36 ins; 5 del; 18 mod
  Patch: https://git.openjdk.org/jdk/pull/15733.diff
  Fetch: git fetch https://git.openjdk.org/jdk.git pull/15733/head:pull/15733

PR: https://git.openjdk.org/jdk/pull/15733