BIO-714: expose + document MXFP8 TriMulPrecision in OSS frontend

[jomitchellnv]

Mirror MXFP8/BFLOAT16 into the fallback TriMulPrecision enum so precision=TriMulPrecision.MXFP8 resolves whether or not the cuequivariance_ops_torch backend is installed. Document the MXFP8 option in the triangle_multiplicative_update docstring as opt-in, inference-only, seq % 32 == 0, Blackwell-only (cc >= 10.0), with transparent bf16 fallback. Add frontend tests: an always-on enum-member check and a GPU-guarded MXFP8 inference exercise (string + enum paths).

Default stays precision=None (off); MXFP8 backward is not yet implemented.

Description

Closes #

Checklist

Please confirm the following before requesting review. PRs that do not meet these requirements cannot be merged. See CONTRIBUTING.md for the full contribution rules.

Contribution rules

• I have read and am adhering to the Contribution Rules.
• An issue was filed and approved by the NVIDIA team before this PR.
• I have installed and run pre-commit locally (pip install pre-commit && pre-commit install).
• This PR addresses a single concern and avoids unnecessary complexity or commented-out code.
• If this PR is a work in progress, the title is prefixed with [WIP].

Sign-off (DCO)

Per Signing Your Work, every commit in this PR must be signed off with git commit -s (appending Signed-off-by: Your Name <your@email.com>). PRs containing unsigned commits will not be accepted.

• All commits in this PR are signed off (git commit -s).
• By signing off, I certify that I have the right to submit this contribution under the project's open source license, in accordance with the Developer Certificate of Origin v1.1.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.