Initial optimization for generating plots with large number of lines

[fbordignon]

This PR attempts to integrate an initial optimization to allow for 1000s of line plots

Fixes #1039

[victorsapucaia]

I used this code snippet to test a solution for issue #1039

import pyqtgraph as pg
import numpy as np
from pyqtgraph.Qt import QtGui

lines = 10
Lx = 64
xs_short = np.arange(Lx)
h_fill = np.zeros(lines)
ys_short = np.zeros((lines,Lx))
for i in np.arange(lines):
ys_short[i,:] = np.sin(xs_short) + i
h_fill[i] = i*1.0

win1 = pg.plot()

pen = pg.mkPen(color=(255, 0, 0))
pen1 = pg.mkPen('b', width=5)

win1.plot(xs_short, ys_short,pen=pen)
win1.plot(xs_short, ys_short,fillLevel=h_fill,brush=(255, 0, 0),fillOutline=False)
win1.plot(xs_short, ys_short,fillLevel=h_fill,brush=(255, 0, 0),fillOutline=True)
win1.plot(xs_short, ys_short[0,:],fillLevel=h_fill[0],brush=(255, 0, 0))
win1.plot(xs_short, ys_short)
p = win1.plot(xs_short, ys_short, pen=pen, symbol='+',)
p = win1.plot(xs_short, ys_short, pen=pen, symbol='+',fillLevel=h_fill,brush=(255, 0, 0),fillOutline=False)
p = win1.plot(xs_short, ys_short[0,:], pen = pen, symbol ='o', symbolPen ='b', symbolBrush = 0.2, name ='blue')
p = win1.plot(xs_short, ys_short, pen = pen, symbol ='o', symbolPen ='b', symbolBrush = 0.2, name ='blue')
p.setShadowPen(pen1)

fillopts = dict(pen=pen, fillLevel=0, fillOutline=True, brush='b')
win1.plot(xs_short, ys_short[0,:], **fillopts)
fillopts = dict(pen=pen, fillLevel=h_fill, fillOutline=True, brush='b')
win1.plot(xs_short, ys_short, **fillopts)