Geometries with fill have visible gaps

Some of the plots don't look very appealing:

```python
import numpy as np
from lets_plot import *
LetsPlot.setup_html()

np.random.seed(43)
data = {
'x': np.append(np.random.normal(0,1,1000), np.random.normal(3,1,500)),
'y': np.append(np.random.normal(0,1,1000), np.random.normal(3,1,500))
}
ggplot(data, aes('x', 'y')) + geom_density2df(aes(fill = '..level..'))
```

```python
import pandas as pd
from lets_plot import *
LetsPlot.setup_html()

df = pd.read_csv("https://raw.githubusercontent.com/JetBrains/lets-plot-docs/master/data/mpg.csv")

n = 12
quantiles=[i/n for i in range(0, n)]

ggplot(df, aes('hwy', group='drv')) + \
geom_density(aes(fill='..quantile..'), quantiles=quantiles, alpha=.97, show_legend=False) + \
ggsize(800, 300)
```

```python
import math
from lets_plot import *
LetsPlot.setup_html()

X_max = 50
Y_max = 50
def z_fun(x, y):
z = math.sin(x * 3 * math.pi / X_max)
z += math.sin(y * 3 * math.pi / Y_max)
z += x * 3 / X_max
z += y * 5 / Y_max
return z

x = []
y = []
z = []

for row in range(0, Y_max - 1):
for col in range(0, X_max - 1):
x.append(col)
y.append(row)
z.append(z_fun(col, row))

dat = dict(x=x, y=y, z=z)

ggplot(dat) + geom_contourf(aes('x', 'y', z='z', fill='..level..'), bins=20)
```

We can try to fix this by using an integer grid. If the coordinate is next to an occupied cell - use the coordinates of the cell, otherwise round the coordinate and take the place in the grid.

Reproduce the visible gaps with the provided geom_density2df, geom_density, and geom_contourf examples. Start by tracing how filled geometry coordinates are generated and rasterized, then evaluate the proposed integer-grid handling for occupied neighboring cells. Done means the supplied filled plots render without visible gaps.

Scritto dal modello di indicizzazione a partire dal testo della issue.