Only measure and store filament width when E is going forward.

Following up on #3231

Sorting ot what else belongs to what if was not so hard.
But the static `ms = ms = millis();` was a bit surpising.

See: https://github.com/Ultimaker/Marlin/pull/7

Apply `constrain`, `NOMORE`, `NOLESS` and `CRITICAL_SECTION` macros
wherever possible.

Redo of #2939 rebased against RCBugFix

st_set_position takes long
```void st_set_position(const long& x, const long& y, const long& z, const long& e)```

anything is long, we don't need float there

dx -> dsx
dy -> dsy
dz -> dsz
de -> dse

Keep "astyled" reformatting

As noted by @darkjavi et.al. in #103, the E1 motor was not being enabled in planner.cpp when using DUAL_X_CARRIAGE. This patch enables and disables E1 as it should.

With the introduction of the #if ENABLED(…)

reads better than

Apply `ENABLED` / `DISABLED` macros to planner files.

Elsewhere DRYRUN turns off the heating elements
and ignores constraints on them.

Here, whenever motion is entered into the planner,
if DRY RUN is set, we instantly act as if the E_AXIS
is in the desired final position.

planner.cpp: single evil and hidden tab removed :recycle: :smirk:

@CONSULitAS

planner.cpp: Add FAN_MIN_PWM for slow fan stopping issue (V2.1 with
macro and linear scaling)

@thinkyhead thanks for idea on linear scaling

replace extruder_multiply with extruder_multiplier
like feedrate_multiplier or volumetric_multiplier

With these changes the output of `M503 S0` is all you need to restore
the EEPROM. Building on this it is straightforward to save and restore
the EEPROM state using the SD card or external GCode file.

- Added `M145` to set “heatup states” for the LCD menu
- Added `M420` to toggle Mesh Bed Leveling
- Added `M421` to set a single Mesh coordinate
- Extended `Config_PrintSettings` with added M codes
- Cleaned up some comments here and there

- Set `de` to 0 when `position[E] = target[E]`
- Address regression mention in #1846