Add the ability to cache large downloads

[lavocatt]

We can enable caching for bash chunks.
The idea would be:

• if cache=true
  • hash the content of the chunk
  • get the folder in which it executed and create a tar archive
  • when encountering a same chunk having a similar hash
  • restore the archive in the directory before executing the chunk

options with wget can then be used to ignore already downloaded files

cache directory can be in ~/.cache/markdown-runner

[gtully]

or a usage pattern, for chunks that download files, use a shared working directory, and wget --N or --timestamping.
we use and document this pattern.

[lavocatt]

I don't know if it's simpler or not to match a pattern yet, let's see what I can come up to.