Improve arrow-ipc documentation

[alamb]

Which issue does this PR close?

• Part of Improve Arrow-IPC performance by avoiding Unsafe Unchecked IPC Read RecordBatch #3287

Rationale for this change

The arrow-ipc docs are quite sparse and don't have many examples. Let's fix that with some documentation TLC

What changes are included in this PR?

1. Add more docs and some basic examples
2. Fix some typos

Are there any user-facing changes?

Just docs, no functional changes

[tustvold]

The Stream and File formats are actually different formats... See https://arrow.apache.org/docs/format/Columnar.html#ipc-file-format

[alamb]

The Stream and File formats are actually different formats... See https://arrow.apache.org/docs/format/Columnar.html#ipc-file-format

Yes, I am not sure if you are suggesting changes to this PR

https://arrow.apache.org/docs/format/Columnar.html#ipc-file-format says:

IPC File Format

We define a “file format” supporting random access that is an extension of the stream format.

[tustvold]

I found the emphasis on the requirement or not for Seek as a way to distinguish the two to be a little backwards. They're different formats, with one allowing random access, the requirement for Seek to be honest is really an implementation detail. FileDecoder decodes the file format, and does not require seek.

[alamb]

I found the emphasis on the requirement or not for Seek as a way to distinguish the two to be a little backwards. They're different formats, with one allowing random access, the requirement for Seek to be honest is really an implementation detail. FileDecoder decodes the file format, and does not require seek.

This is a good point. I tried to clarify in a335701