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

The match/matches methods of UriTemplate use a regex with (.*) in place
of URI variables, which work fine except in the end where such a
pattern can match greedily more than one segment.

This commit updates the regex to use ([^/]*) instead since URI
variables are only meant to be used within a single path segment.

Issue: SPR-16169

Clarify what PathPatternParser is used for which is CORS checks in the
very least. Some sub-classes will also use it for request mapping but
not all (e.g. RouterFunctionMapping). Hence the need to be more
explicit.

@EnableWebFlux bootstraps both annotated controllers and functional
endpoints, so we need to be more explicit about which parts of the
configuration apply to which.

Issue: SPR-16360

Issue: SPR-16329

Issue: SPR-16288

Issue: SPR-16296

Issue: SPR-16360

This commit polishes SPR-16022 fix in order to handle correctly
the case when primary and default constructors are the same when
a secondary constructor is defined.

Issue: SPR-16289

Reverted back from `MultiValueMap<String, Object>` to
`MultiValueMap<String, ?>`

Issue: SPR-16246

This commit uses the newly introduced write aggregator from
DataBufferUtils throughout WebFlux.

Issue: SPR-16351

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

Prior to this commit, NettyDataBuffer had a optimization in
write(ByteBuf...), where it used a CompositeByteBuf to hold the original
and the parameter buffer.
Unfortunately, this procedure has nasty consequences when splicing
buffers (see https://stackoverflow.com/a/48111196/839733).

As of this commit, NettyDataBuffer stopped using CompositeByteBuf, and
 simply does ByteBuf.write().

Issue: SPR-16351

Issue: SPR-16290

Issue: SPR-16245

Includes default implementations of getSqlType and getTypeName.

Issue: SPR-16361

Includes efficient separator determination.

Issue: SPR-16340

Issue: https://jira.spring.io/browse/SPR-16235

* pr/1640:
  Fix source code in doc

Closes gh-1640

Optimize same origin check when the request is an instance of
ServletServerHttpRequest and when there is no forwarded headers.

This commit also optimizes the getPort methods and ForwardedHeaderFilter
forwarded headers checks.

Issue: SPR-16262

This commit refines forwarded protocol support in order to support
proxies that only set "X-Forwarded-Proto" header and not
"X-Forwarded-Port" by performing a reset of the port in such case.

"Forwarded" header support has been updated accordingly since it
also supports similar use case, as described in SPR-15504.

Issue: SPR-16262

Issue: SPR-16214

Issue: SPR-16324

Issue: SPR-16322

Issue: SPR-16266

Issue: SPR-16356

Issue: SPR-16315

Issue: SPR-16353

Issue: SPR-16352