feat: make onnx export a function

[stephantul]

This PR makes the ONNX export we offered a part of the model2vec package. I added extra things to the onnx extra to support new torch versions and made the script work on an in-memory version of a model, not one stored on disk.

from model2vec import StaticModel
from model2vec.onnx import export_model_to_onnx

s = StaticModel.from_pretrained("minishlab/potion-base-8m")
export_model_to_onnx(s, "my_model")

This writes an onnx model to disk, which can then be uploaded to Hugging Face manually and used wherever. I chose to write some additional config options which were needed by fastembed. Most of these are not necessary, which I reported here.

[codecov]

Codecov Report

❌ Patch coverage is 98.38710% with 2 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
model2vec/onnx.py	98.36%	2 Missing ⚠️

Files with missing lines	Coverage Δ
model2vec/tokenizer/tokenizer.py	100.00% <100.00%> (ø)
model2vec/onnx.py	98.36% <98.36%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.