release/1.9.0: bug fixes for fckit and ectrans, set BLA_VENDOR in eckit

[climbfuji]

Summary

This PR updates the submodule pointer for spack for the changes in:

• release/1.9.0: bug fixes for fckit (disable finalization of DDTs) and ectrans (disable use of Fortran contiguous keyword) spack#518
• eckit: set cmake BLA_VENDOR spack#503

Testing

• Built unified environment on @climbfuji's laptop with [email protected] (no testing)
• CI
• TESTING WITH JEDI (@fmahebert)

Applications affected

JEDI

Systems affected

All systems with compilers that ectrans and fckit didn't handle correctly

Dependencies

• waiting on release/1.9.0: bug fixes for fckit (disable finalization of DDTs) and ectrans (disable use of Fortran contiguous keyword) spack#518

Issue(s) addressed

Closes #1521
Closes #1522

Checklist

• This PR addresses one issue/problem/enhancement, or has a very good reason for not doing so.
• These changes have been tested on the affected systems and applications.
• All dependency PRs/issues have been resolved and this PR can be merged.

[fmahebert]

Testing this now... will take a bit of time though, as you well know!

[climbfuji]

Testing this now... will take a bit of time though, as you well know!

Should I wait with merging the spack PR (not this spack-stack PR) until your testing is done?

[fmahebert]

I should have results (one way or another) by end of day, so perhaps let's hold off on merging...

[fmahebert]

On my Ubuntu24 env with gcc13, I've built a new spack-stack from your fork and this branch. On this stack, jedi-bundle builds and ctests pass with no unexpected failures (one known failure remains).

Thus, I think we're good to proceed.

Thanks @climbfuji for the rapid iteration!