More CSV statistics helpers

[Notgnoshi]

Closes #109 Closes #110 Closes #111 Closes #112 Closes #106 Closes #117

Use case: Given a candump filtered to a specific message, understand the latency distribution.

$ sudo ip link add dev can0 type vcan
$ sudo ip link set up can0
$ cangen can0 &
$ candump -l can0
# wait
$ can2csv candump.log >candump.csv
$ csvdelta -c timestamp --inplace candump.csv
$ csvstats -c "timestamp deltas" candump.csv --histogram 3
num: 539
min: 0.20007610321044922 index: 103
max: 0.6002910137176514 index: 199
sum: 117.06238389015198
mean: 0.21718438569601473
stddev: 0.0572751643876145

This isn't a very good example because cangen doesn't have enough variance. Here's a real one

Also note that if you don't specify the number of histogram bins yourself, seaborn took way too long (I gave up) to calculate an appropriate number of bins.

[Notgnoshi]

What if instead of a histogram, I make a timeseries plot with the timestamps as the x-axis, and the time-since-last message as the y-axis?

$ csvplot -y time-since-last-das -y time-since-last-sas --ylim 0 0.2 --xlim 20000 60000  sas.csv