Speed test + largemix mpnn config

[zhiyil1230]

Changelogs

• enumerate the changes of that PR.

---

Checklist:

• Was this PR discussed in an issue? It is recommended to first discuss a new feature into a GitHub issue before opening a PR.
• Add tests to cover the fixed bug(s) or the new introduced feature(s) (if appropriate).
• Update the API documentation is a new function is added, or an existing one is deleted.
• Write concise and explanatory changelogs above.
• If possible, assign one of the following labels to the PR: feature, fix or test (or ask a maintainer to do it for you).

---

discussion related to that PR

• optimum parameters for the training speed of largemix and pcqm4m (mpnn): 100 epochs, 14 hours for largemix gcn and 8 hours for pcqm4mv2 mpnn
• add mpnn largemix config (currently not in hydra, will convert to it once the largemix conversion is ready)

[codecov]

Codecov Report

Merging #424 (2d51bbb) into main (57399b3) will decrease coverage by 1.66%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##             main     #424      +/-   ##
==========================================
- Coverage   66.91%   65.25%   -1.66%     
==========================================
  Files          82       82              
  Lines        7819     7806      -13     
==========================================
- Hits         5232     5094     -138     
- Misses       2587     2712     +125     

Flag	Coverage Δ
unittests	65.25% <ø> (-1.66%)	⬇️

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

Components	Coverage Δ
ipu	48.75% <ø> (-0.40%)	⬇️

[DomInvivo]

@zhiyil-graphcore , can you resolve the conflicts that were caused by merging the last 2 PR's #421 and #423 ? I'm afraid of breaking something if I do it myself.

[zhiyil1230]

@zhiyil-graphcore , can you resolve the conflicts that were caused by merging the last 2 PR's #421 and #423 ? I'm afraid of breaking something if I do it myself.

I've made the changes accordingly with the latest change on hydra configs.