FIX: Update env var settings for multi node

[mebristo]

Changes in the way that Azure ML set the environment variable ENV_AZ_BATCHAI_MPI_MASTER_NODE mean that the logic for setting environment variables for multi-node jobs needs updating

[codecov[bot]]

Codecov Report

Merging #588 (1d163bf) into main (1efdced) will decrease coverage by 0.17%. The diff coverage is 7.40%.

Flag	Coverage Δ
hi-ml	81.73% <0.00%> (-0.08%)	:arrow_down:
hi-ml-azure	25.26% <8.00%> (-0.21%)	:arrow_down:
hi-ml-cpath	77.39% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
hi-ml/src/health_ml/runner.py	80.29% <0.00%> (-1.33%)	:arrow_down:
hi-ml-azure/src/health_azure/utils.py	25.49% <8.00%> (-0.39%)	:arrow_down:

[fepegar]

LGTM, but some tests are failing.

Also, @Shruthi42 is probably a much more appropriate reviewer than I am for this PR.

[mebristo]

LGTM, but some tests are failing.

Also, @Shruthi42 is probably a much more appropriate reviewer than I am for this PR.

Good to know, thanks!

[mebristo]

Looking good! I added suggestions for corner cases. Also, you had another safety check in the runner, to only call the function when we are really doing multi-node? Should we add a multi-node training job to the hi-ml test suite in a follow-up PR (happy to work on that myself)?

Good idea to add the multi-node training job in a follow-up PR