Document how to size the direct memory under expected event size

[andsel]

Release notes

[rn:skip]

What does this PR do?

Better document how this plugin allocates direct memory so that a better capacity planning can be applied.

Why is it important/What is the impact to the user?

The user, given some statistical measures, can determine if the sizing of direct memory is correct so that he can avoid potential out-of-memory errors in the direct memory.

Checklist

• My code follows the style guidelines of this project
• [ ] I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• [ ] I have made corresponding change to the default configuration files (and/or docker env variables)
• [ ] I have added tests that prove my fix is effective or that my feature works

[andsel]

Note for reviewer:
Beats decoding part keeps 2 copies of the buffer it's processing in memory:

• one for the decompression:

  logstash-input-beats/src/main/java/org/logstash/beats/BeatsParser.java

  Lines 232 to 234 in 339913c

  	ByteBufOutputStream buffOutput = new ByteBufOutputStream(out);
  	InflaterOutputStream inflaterStream = new InflaterOutputStream(buffOutput, inflater)
  	) {

• the other is kep in the batch before lazily creating LogStash events:

  logstash-input-beats/src/main/java/org/logstash/beats/V2Batch.java

  Line 12 in 339913c

  private ByteBuf internalBuffer = PooledByteBufAllocator.DEFAULT.buffer();