Codechange: Make all NWidgetPart arrays constexpr.

This ensures that the arrays are not created at runtime and prevents using non-constexpr values.

Codechange: Replace mishmash of types for widget index with WidgetID.

Indices were stored as int, but often passed around as uint/uint8_t and casts.

Now they should all use WidgetID.

Change: Use CRTP-mixins to compose dropdown list items.

This allows list items to built from component parts as required, and additional
functionality is added:

* Icons and text can be positioned at the start or end of the space (templated.)
* Font size of text can be changed (templated.)
* Palette of sprites can be set (runtime.)

Change: Add extra padding to climate buttons to match pre-#11464. (#11525)

Padding used to be included in the SetMinimalSize() part which was removed, but also made it require specific sprite sizes.

This now adds padding on the already determined size, removing the need for hardcoding pixel dimensions and allowing the sprites to be any size.

Change: Set height of larger heightmap/scenario worldgen buttons by text height. (#11496)

This allows the button height to scale properly by font size.

Codechange: Remove FONT_HEIGHT_... macros. (#11481)

These make it look like we're dealing with a constant, but actually each is a call to `GetCharacterHeight(...)`.

Codechange: Add base() method to StrongType to allow access to the base type without casting. (#11445)

This removes the ability to explicitly cast to the base type, but the requirement
to use .base() means the conversion is still explicit.

Codechange: Add __FILE__/__LINE__ of WindowDesc to WindowDesc.

This is to allow unit-tests to produce more useful output.

Codechange: Avoid emplace_back(new()) into a unique_ptr. (#11384)

This could theoretically leave an unmanaged pointer in certain circumstances, and directly using
make_unique shows intent.

Codechange: Use begin/end of nwidget parts of begin/length.

This simplifies processing nwidget parts as, unlike the remaining length, the pointer to the end of the list never changes. This is the same principle as we use(d) for tracking end instead of length for C-style strings.

And this removes 160~ instances of the lengthof() macro.

Codechange: automatic adding of _t to (u)int types, and WChar to char32_t

for i in `find src -type f|grep -v 3rdparty/fmt|grep -v 3rdparty/catch2|grep -v 3rdparty/opengl|grep -v stdafx.h`; do sed 's/uint16& /uint16 \&/g;s/int8\([ >*),;[]\)/int8_t\1/g;s/int16\([ >*),;[]\)/int16_t\1/g;s/int32\([ >*),;[]\)/int32_t\1/g;s/int64\([ >*),;[]\)/int64_t\1/g;s/ uint32(/ uint32_t(/g;s/_uint8_t/_uint8/;s/Uint8_t/Uint8/;s/ft_int64_t/ft_int64/g;s/uint64$/uint64_t/;s/WChar/char32_t/g;s/char32_t char32_t/char32_t WChar/' -i $i; done

Codechange: Make DropDownListStringItem preformat and remove other implementations. (#11063)

Having to choose between DropDownListStringItem, DropDownListCharStringItem, and DropDownListParamStringItem depending on whether to draw a StringID, a raw string, or a StringID with extra parameters was needlessly complex.

Instead, allow passing a StringID or raw string to DropDownListStringItem. This will preformat the StringID into a raw string, and can therefore accept parameters via the normal SetDParam mechanism.

This also means that strings no longer need to be formatted on every draw.

Change: Rename text colour NWidgetPart to SetTextStyle and add font size.

This allows to set both text colour and size for any widget with default
drawing.

Fix: Don't double-pad existing size when adding widget padding.

UpdateWidgetSize's size parameter already includes padding.

Change: Standardize progress bar layout.

Progress bars are drawn differently depending on when it was added, with
different layouts and sizes.

This change adds a standard padding size to use, and makes all progress
bars visually similar, with scaled padding.

Change: Use Rect helpers for widget drawing.

This replaces repetitive and sometimes unwieldy use of constants.

Cleanup: Text widgets are capable of drawing text without assistance.

Remove custom text drawing of some widgets in favour of standard text
widgets.

Change: Don't use zero-sized plane for World Generation window.

Use of zero-sized plane causes the window size calculation to change
depending on which plane is displayed. Instead use an empty plane so
that largest of the planes is taken into account for sizing.

Change: reworked the debug levels for network facility (#9251)

It now follows very simple rules:
0 - Fatal, user should know about this
1 - Error, but we are recovering
2 - Warning, wrong but okay if you don't know
3 - Info, information you might care about
4 -
5 - Debug #1 - High level debug messages
6 - Debug #2 - Low level debug messages
7 - Trace information

Feature: allow setting a custom terrain type to define highest peak

At least, TGP will try to reach it. It heavily depends on the map
if it is reachable at all. But for sure it will do its atmost to
get there!

Add: allow setting the highest mountain for heightmaps

It will add some slack to the map height limit if that was set
to auto.

Feature: auto-detect map height limit based on generated map

This opens up the true power of the TGP terrain generator, as it
is no longer constrainted by an arbitrary low map height limit,
especially for extreme terrain types.

In other words: on a 1kx1k map with "Alpinist" terrain type, the
map is now really hilly with default settings.

People can still manually limit the map height if they so wish,
and after the terrain generation the limit is stored in the
savegame as if the user set it.

Cheats still allow you to change this value.

Change: rename setting "max_heightlevel" to "map_height_limit"

This better reflects what it is, and hopefully removes a bit of
the confusion people are having what this setting actually does.

Additionally, update the text on the setting to better inform
users what it is doing exactly, so they can make an educated
decision on how to change it.

Next commit will introduce an "auto" value, which should be the
new default. The rename has as added benefit that everyone will
start out on the "auto" value.

Change: remove "maximum map height" from the New Game GUI

This setting influence the max heightlevel, and not as the name
suggests: the height of the generated map.

How ever you slice it, it is a very weird place to add this
setting, and it is better off being only in the settings menu.

Commits following this commit also make it more useful, so users
no longer have to care about it.

Feature: setting to indicate desert coverage for tropic climate

This is an indication value; the game tries to get as close as it
can, but due to the complex tropic rules, that is unlikely to be
exact.

In the end, it picks a height-level to base the desert/tropic
line on. This is strictly seen not needed, as we can convert any
tile to either. But it is the simplest way to get started with
this without redoing all related functions.

Feature: setting to indicate snow coverage for arctic climate (replaces snow line height)

Setting the snow coverage (in % of the map) makes a lot more sense
to the human, while still allowing the niche player to set (by
finding the correct %) a snow line height they like. This makes for
easier defaults, as it decoupled terrain height from amount of snow.

Maps can never be 100% snow, as we do not have sprites for coastal
tiles.

Internally, this calculates the best snow line height to approach
this coverage as close as possible.

Fix: placing random trees in SE crashes the game (#8892)

This used to work by accident: originally the code checked if
GenerateWorld was threaded. If not, it would abort the function.
This worked for placing trees, because it was also returning false
when it was not active.

With the recent changes, that check got removed, and this crash
started to happen. So now check if we have a modal window, which
is a very solid indication we are generating the world.

Add: make modal windows update more smooth

Basically, modal windows had their own thread-locking for what
drawing was possible. This is a bit nonsense now we have a
game-thread. And it makes much more sense to do things like
NewGRFScan and GenerateWorld in the game-thread, and not in a
thread next to the game-thread.

This commit changes that: it removes the threads for NewGRFScan
and GenerateWorld, and just runs the code in the game-thread.
On regular intervals it allows the draw-thread to do a tick,
which gives a much smoother look and feel.

It does slow down NewGRFScan and GenerateWorld ever so slightly
as it spends more time on drawing. But the slowdown is not
measureable on my machines (with 700+ NewGRFs / 4kx4k map and
a Debug build).

Running without a game-thread means NewGRFScan and GenerateWorld
are now blocking.

Fix d437445c: also use std::chrono for the GRFFileScanner modal window

For some reason I only converted one of the two modal windows we
have, and completely forgot the other.

While at it, synchronize the way those two modal windows work
in terms of "next_update".

Codechange: use std::chrono to track time in modal windows

Adding to _realtime_ticks in a random place is a bit of a hack,
and by using modern C++, we can avoid this hack.

Codechange: Switch DropDownList to directly use std::vector, thus making AutoDeleteSmallVector obsolete.

DropDownListItem are strongly managed using std::unique_ptr to ensure leak-free handling. Appropriate use
of move-semantics make intent a lot clearer than parameter comments and allows the compiler to generate
copy-free code for most situations.

Codechange: Replace custom mutex code with C++11 mutex'es.

A conforming compiler with a valid <mutex>-header is expected.
Most parts of the code assume that locking a mutex will never fail unexpectedly,
which is generally true on all common platforms that don't just pretend to
be C++11. The use of condition variables in driver code is checked.

Fix #6898: Replace atoi() with strtoul()

Normalize type and parsing of generation_seed across all files
Add assert_compile() to ensure correct type