relax glob pattern to find Mathematica install script

[boegel]

(created using eb --new-pr)

For Mathematica 13.1.0, the name of the script is just Mathematica_13.1.0_LINUX.sh

[Micket]

lgtm (though i don't have access to that version yet)

[boegel]

Test report by @boegel

Overview of tested easyconfigs (in order)

• FAIL (build issue) Mathematica-12.1.1.eb (partial log available at https://gist.github.com/85cee6d3935d836a2b4b12e191107b90)
• FAIL (build issue) Mathematica-13.0.0.eb (partial log available at https://gist.github.com/a543998f5efe7172132ef69647af72b6)

Build succeeded for 0 out of 2 (2 easyconfigs in total)
node3704.slaking.os - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz (haswell), Python 3.6.8
See https://gist.github.com/0ec325e1bc910395b6e8529d56d63ff2 for a full test report.

[verdurin]

@boegel I'm afraid I merged the easyconfig before @Micket pointed out that this easyblock hadn't gone in yet.

[boegel]

I'm not sure what's causing the Segmentation fault when running mathematica --version during the sanity check, but I guess it could have something to do with not having enough memory available.

It definitely has nothing to do with the changes being made here to the easyblock though, that's clear imho...

[boegel]

Test report by @boegel

Overview of tested easyconfigs (in order)

• SUCCESS Mathematica-12.1.1.eb
• SUCCESS Mathematica-13.0.0.eb
• SUCCESS Mathematica-13.1.0.eb

Build succeeded for 3 out of 3 (3 easyconfigs in total)
node3121.skitty.os - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz (skylake_avx512), Python 3.6.8
See https://gist.github.com/5e19e102b1601adf8cbd9ae8ff589413 for a full test report.