I am not sure if there is a rationale behind ignoring extra keys like
this. I was under the impression that something might copy options to
platform specific options, thus leaving options in the original array,
knowing they would be ignored anyway. If it is the case, it would be
best to just move them, otherwise, this commit might help troubleshoot
some mapping issues.
I had to fix some tests that wrongly used that options array (maybe
these options were valid at some point?).

It makes diffs easier to read and avoids conflicts.

Refactoring tests

Fixes for Sqlite schema regex

The reasoning behind this is to essentially rely on a single type (or
the absence of it), therefore add type declarations.

Since these are private methods we can do that without breaking BC.

Fix changing column from int to bigint with autoincrement

- SERIAL and BIGSERIAL are not true types and can't be used in ALTER TABLE expressions
- When changing between int and bigint auto increment fields the default should no be dropped
  to preserve the link between the column and the sequence.

Inherit charset from master connection if not set explicitly

added badges for continuousphp and other badges for 2.6

Also reorganising things a bit to make it simpler.

MariaDB 10.2 initial support

Fixes: https://github.com/doctrine/dbal/issues/2819

As per @belgattitude's review.

Since the test is skipped when the driver is not an instance of
`VersionAwarePlatformDriver` we don't need to validate it again.

The message was a bit confusing as well (expected vs given).

Since the JSON type is simply an alias to a LONG TEXT we must force
a comment, otherwise things will not work properly.

Also the test for JSON types should work regardless of native JSON
support (DBAL should add a comment to identify the correct type).

Unescaping re-added, support limited to schema introspection (#2850 will bring more complete coverage)