Add extra links to README and improve description

[matt-graham]

Adds badges to README for latest stable version of documentation and GMD journal article (will need to be updated once the accepted version is available online). Also tweaks the description a little bit to mention thread-based parallelism and add a link to ARC website. I also noticed there was a unit_test.h5 HDF5 file we don't appear to use anywhere that was added in #218, I assumne accidentally, so have removed this.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.48%. Comparing base (3619473) to head (5291fda).
Report is 3 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #264   +/-   ##
=======================================
  Coverage   92.48%   92.48%           
=======================================
  Files           7        7           
  Lines         386      386           
=======================================
  Hits          357      357           
  Misses         29       29           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[giordano]

I now renamed the branch, so should replace master -> main everywhere. Sorry! 😅

[matt-graham]

Do we also want to change master_rank → main_rank?

[giordano]

Potentially yes, but it can be done in a separate PR. Also, it'd probably be a technically breaking change, as I think we expose the name to users at

ParticleDA.jl/src/params.jl

Line 6 in 1618055

* `master_rank` : Id of MPI rank that performs serial computations

[matt-graham]

Looks like there is some implicit reference to master branch in benchmarks workflow from the fact it fails with

fatal: couldn't find remote ref refs/heads/master

when running

using PkgBenchmark, BenchmarkCI

BenchmarkCI.judge(
    PkgBenchmark.BenchmarkConfig(
        env = Dict(
            "JULIA_NUM_THREADS" => "2",
        ),
    ),
)

but not sure how to fix - @giordano any ideas?

EDIT: Maybe just passing id="main" to PkgBenchmark.BenchmarkConfig will fix?

(Second) EDIT: Looks like culprit is baseline = "origin/master" default in BenchmarkCI.judge so will try updating that.