Make `rascal.toml` an alternative to `rascal.mf`

[DavyLandman]

This would be an example rascal.toml

version='1.0' 
[project]
name = 'test-project'
sources = [ 'src/main/rascal']
libraries = [ '|lib://typepal|']
[main]
module = 'lang::testing::Mod'
function = 'main'

[codecov[bot]]

Codecov Report

Merging #1881 (85b8a77) into main (7f48fd7) will decrease coverage by 1%. The diff coverage is 0%.

@@           Coverage Diff           @@
##              main   #1881   +/-   ##
=======================================
- Coverage       49%     49%   -1%     
  Complexity    6155    6155           
=======================================
  Files          660     661    +1     
  Lines        58668   58716   +48     
  Branches      8543    8548    +5     
=======================================
+ Hits         28916   28920    +4     
- Misses       27559   27606   +47     
+ Partials      2193    2190    -3     

Files	Coverage Δ
...rc/org/rascalmpl/interpreter/utils/RascalToml.java	0% <0%> (ø)

... and 5 files with indirect coverage changes

[jurgenvinju]

I have sympathy for this because RASCAL.MF syntax is so finicky, but let's not do this. Reasons:

• Not yet another format; we don't even support this one yet with a reader/writer in the library. Where we do have XML, JSON and MF and YAML support.
• We are moving anyway towards depending purely and solely on pom.xml for configuring projects and META-INF for information about deployed artefacts (just like pure Java projects). This would remove the need for RASCAL.MF entirely, which was invented to bridge the gap between commandline/maven projects and Rascal Eclipse projects without Maven pom.xml. If all Rascal projects have a pom.xml file, then it is not needed anymore. The MF format was not the right choice for configuring source projects; we should have gone with XML or TOML files right from the start.