This change improves the report from benchmark runs by adding t-tests
for each scenario. This allows the benchmarks to provide a summary at
the end with the scenarios that are likely to show perfomance difference
between the latest released version and the one that comes with the PR
the benchmarks run for.

This also updates the versions that are being benchmarked to include the
latest releases.

ci(benchmarks): improve benchmark reporting

This change adds a data validation workflow. Data is validated by
performing a Hotelling T2 test on the collected data.

ci: add data validation workflow

This change adds extra steps to the CI workflows to build Python
wheels so that the Austin binary can be installed with pip from
PyPI.

ci: build Python wheels

Unused variable alloc

The d_name variable is a table of 256. It can not be null.

Use the more correct expression for computing the lasti index with
Python 3.11.

ci(ubuntu): update before install

The version of cppcheck available from the package manager tends to be
outdated. In this change we build from sources instead. We cache the
build result for faster execution.

chore: minor compilation fixes with gcc 11.3

ci(checks): build cppcheck from sources

refactor: use proper lasti formula for 3.11