This commit fixes an issue where DataBufferUtils.join() would not
release databuffers that preceded an error signal.

Issue: SPR-17025

Issue: SPR-17188

Fix checkstyle violations for javadoc.

Issue: SPR-16968

(cherry picked from commit a2765c00)

Issue: SPR-16337

Issue: SPR-16365

Fixed by creating `Callable`-based variants, as explained in the JIRA
issue.

Issue: SPR-16403

Use DataBufferUtils.compose instead of writeAggregator to combine
multiple data buffers into one, as the write aggregator would not work
when the initial data buffer did not have enough capacity to contain
all subsequent buffers.

Removed writeAggregator, as it is no longer needed.

Issue: SPR-16365

Added a utility method that composes data buffers into a single buffer.
Depending on the `DataBuffer` implementation, the returned buffer may be
a single buffer containing all data of the provided buffers, or it may
be a true composite that contains references to the buffers.

Issue: SPR-16365

As a consequence of dropping CompositeByteBuf (see prior commit),
DataBuffers fluxes that are aggregated with Flux.reduce(BiFunction) are
now required to be released, as the composite no longer holds a
reference to subsequent data buffers.

For this purpose, DataBufferUtils now has a writeAggregator that can be
used with Flux.reduce, and that released the subsequent buffers
properly.

Issue: SPR-16351

This commit introduces various improvements in DataBuffer:

- DataBuffer now exposes its read and write position, as well as its
capacity and writable byte count.
- Added DataBuffer.asByteBuffer(int, int)
- DataBufferUtils.read now reads directly into a DataBuffer, rather than
copying a ByteBuffer into a DataBuffer
- TomcatHttpHandler now reads directly into a DataBuffer

Issues: SPR-16068 SPR-16070

This commit fixes an issue in the DataBufferUtils.write variant that
takes a AsynchronousFileChannel.

Issue: SPR-15798

Currently ResourceEncoder and ResourceRegionEncoder use DataBufferUtils
to read resource with an AsynchronousFileChannel if it is a file or
otherwise fallback on getting the channel from the resource.

The same is now required in other places where a Resource needs to be
read and is also generally useful.

Issue: SPR-15773

This commit changes the write methods to return `Flux<DataBuffer>`
instead of `Mono<Void>`, giving access to the original buffers,
so that they can decided whether the buffers need to be closed or not.

Issue: SPR-15726

This commit adds an overloaded write method to `DataBufferUtils`. There
are three parameter variants: `OutputStream`, `WritableByteChannel`, and
`AsynchronousFileChannel`.

Issue: SPR-15726

This commits extends nullability declarations to the field level, formalizing the interaction between methods and their underlying fields and therefore avoiding any nullability mismatch.

Issue: SPR-15720

Beyond just formally declaring the current behavior, this revision actually enforces non-null behavior in selected signatures now, not tolerating null values anymore when not explicitly documented. It also changes some utility methods with historic null-in/null-out tolerance towards enforced non-null return values, making them a proper citizen in non-null assignments.

Some issues are left as to-do: in particular a thorough revision of spring-test, and a few tests with unclear failures (ignored as "TODO: NULLABLE") to be sorted out in a follow-up commit.

Issue: SPR-15540

Issue: SPR-15318

Currently the BOM versions are:

* reactor-core 3.0.6.BUILD-SNAPSHOT
* reactor-netty 0.6.2.BUILD-SNAPSHOT

This commit fixes as well a few deprecations in reactor-core.

This commit adds an overloaded DataBufferUtils.read method that operates
on a AsynchronousFileChannel (as opposed to a ReadableByteChannel, which
already existed). This commit also uses said method in the Resource
encoders, if the Resource is a file.

This commit adds the necessary infrastructure for the support of HTTP
Range requests. The new `ResourceRegionEncoder` can write
`ResourceRegion` objects as streams of bytes.

The `ResourceRegionEncoder` relies on an encoding hint
`BOUNDARY_STRING_HINT`. If present, the encoder infers that multiple
`ResourceRegion`s should be encoded and that the provided boundary
String should be used to separate ranges by mime boundaries.
If that hint is absent, only a single resource region is encoded.

Issue: SPR-14664

Fixed bug where the returned Flux from DataBufferUtils.read() would be
completed prematurely if the channel was not ready to read, but did
not reach the end of the file either.

DataSourceUtils moved to main core.io.buffer package.
Consistently named Jackson2JsonDecoder/Encoder and Jaxb2XmlDecoder/Encoder.
Plenty of related polishing.

Ensure type-level Javadoc in every class, comply with guidelines for
80 char on Javadoc, and minor polish.

- Added fromIndex parameter to indexOf and lastIndexOf
- Moved DataBuffer.tokenize to StringEncoder, as that's the only place
  it's used.

This commit introduces DataBuffer.indexOf(IntPredicate) and
lastIndexOf(IntPredicate), for finding the index of a byte in a
DataBuffer.
It also introduces DataBufferUtils.tokenize, which tokenizes a
DataBuffer into separate tokens, given a delimiter function.