Older clang versions raise:
```
./flatbuffers/base.h:365:1: error: unknown attribute 'no_sanitize' ignored [-Werror,-Wattributes]
__supress_ubsan__("alignment")
^
./flatbuffers/base.h:246:50: note: expanded from macro '__supress_ubsan__'
  #define __supress_ubsan__(type) __attribute__((no_sanitize(type)))
                                                 ^
```

Comparing https://releases.llvm.org/3.6.0/tools/clang/docs/AttributeReference.html
with https://releases.llvm.org/3.7.0/tools/clang/docs/AttributeReference.html
shows that __attribute__((no_sanitize(type))) is available since 3.7.0

* Skip generation of mutable union types

* Removed C# and Java unit tests that mutated a Union type

Fixes following clang -Wdocumentation warning:
```
flatbuffers.h:1762:17: error: parameter ']' not found in the function declaration [-Werror,-Wdocumentation]
  /// @param[in]] v A const reference to the `std::vector` of structs to
```

Kolin uses java library as dependency, which changed the way it access union vector recently
(e365c502).

This changes updates kotlin code generation to match Java's changes.

The condition was unnecessary and Detected by

PVS-Studio
V560 [CWE-571] A part of conditional expression is always true: !opts.use_flexbuffers. flatc.cpp 438

* start

* works for current usages!

* unpack: vector of struct

* optimize byte slice

* support nested struct

* support null table

* support struct

* support union

* update generated code

* grumble

* fix compiler warning

* update generated code

* wrap type in namespace

* bug

* wrap in namespace

* enum byte arrays

* generate struct for unions

* basic testing

* remove branching

* fix assert

* pack vector of fixed structs correctly

* omit null vectors

* Refactor Union Pack and UnPack methods

Remove append usage to increase code efficiency when dealing with large vectors

* generate goldens

* Rust: Add idiomatic iterator for Vector type

* Add comments explaining some implementation details

Found by the https://github.com/OSGeo/gdal/blob/master/gdal/scripts/fix_typos.sh
script on the internal copy of flatbuffers inside GDAL

* Rust: Fix Copy and Clone impls for a few generic types

* Add tests for Copy+Clone

* Wrap Copy+Clone checks in a #[test] function

- `--clean-first` runs `clean` target to remove previously generated files
- missed `monster_extra_generated.h` file added to the `flattests` target

- remove not necessary file `monster_test_generated.h` from sample_bfbs.cpp

* Avoid intentional unsigned integer overflow getting caught by sanitizers

* Adapt from suggested compiler compatible solution

* And a third way :)

* Add forceDefaults opt to python Builder

* Add test functions for force_default option for python builder

* Simplify

* Add force default test for UOffsetTFlags

* Annotate getters with @Pure when --java-checkerframework is specified.

Together with @Nullable, this allows users to use static analysis tools
like CheckerFramework to catch NPEs caused by unset fields.

* Don't annotate vector-of-tables item getters with @Nullable.

Since Flatbuffers don't support null items in vectors of tables.

Change-Id: I6fd6283b616c7a39bb878b1610e4ddf6e208fa0a

* byte buffer factory returned buffer is used instead of the requested capacity

* byte buffer factory returned buffer is used instead of the requested capacity

* Comment fix

* Fix C/C++ Create<Type>Direct with sorted vectors

If a struct has a key the vector has to be sorted. To sort the vector
you can't use "const".

* Changes due to code review

* Improve code readability

Change-Id: Iacc0c8513af80a736700e6cbaf513ebdf8e3ac89

In my benchmarks it shows deserialization performance improvements of
around 7%

Empty objects that inherit from Sized would try to access internal
ByteBuffer when Sized::size was called. So we add a single byte in
the empty buffer, so when size() is called it would return 0

Change-Id: I59ff072e526fc63b3215767a4d4a2a8944b65654

Changes the use of `LookupStruct` to `LookupCreateStruct` in
`ParseField` to also detect when collisions happen in namespaces.

Alongside --binary and --json, an additional
switch (--flexbuffers) can be passed to convert
data using FlexBuffers

Fixes build errors from using newer crate versions with an old Rust version.

* Adds XOPEN_SOURCE for PATH_MAX and POSIX 1993 for stat

These are the only two required extension for compilation of
flatbuffers using -std=c++11 instead of gnu++11.

* Sets _XOPEN_SOURCE to 600 and enable POSIX2001 for fseeko

The real position of a string is  calculated by using the indirect() method,
which should be based on parentWidth and not byteWidth, as it was implemented.

We are also fixing the flag BUILDER_FLAG_SHARE_STRINGS on FlexBuffersBuilder
that was set as '1', same value as BUILDER_FLAG_SHARE_KEYS.

* Draft with Array specialization (#5508)

* Array specialization + SFINAE to fold copy-paste (#5508)

* Add implicit specialization of Array<scalar> and Array<struct> (#5508)

- Tag dispatching is used for implicit specialization
- Array<scalar> and Array<struct> have different iterators and accessors
- Array<scalar> and Array<struct> have different Mutate() methods

* Add implicit specialization of Array<scalar> and Array<struct> (#5508)

- Tag dispatching is used for implicit specialization
- Array<scalar> and Array<struct> have different iterators and accessors
- Array<scalar> and Array<struct> have different Mutate() methods

* Java: Added access object for vector of struct and vector of tables.

* Java: Workarounds removed when accessing the union vector.

Fixes warnings caught in:
https://buildkite.com/bazel/flatbuffers/builds/863#cff87e1d-b976-4734-8157-4b6d88c7b5e1

Change-Id: Ie94386ff8e10fd2a964bd9155139b50953746a37

Change-Id: Ie82abaf178495c4692e7d10be6b4a13f2fa1bee6

* Add element size parameter to __vector_as_arraysegment

Add element size parameter to __vector_as_arraysegment fixing issue where VectorAsBytes returns incorrect size span for multibyte element types.

* Update codegen

Update codegen and Table to return typed span.

* update test files

update test files