Warn when using fp8 weights + non-fp8 computation

[kunlunl]

Description

Currently, TE's Linear, LayerNormLinear and GroupedLinear de-quantized the fp8 weights to bf16 when self.fp8 is False. This MR changes the de-quantization to raising an error, to prevent fp8 weights + non-fp8 computation.

Type of change

• [ ] Documentation change (change only to the documentation, either a fix or a new content)
• [x] Bug fix (non-breaking change which fixes an issue)
• [ ] New feature (non-breaking change which adds functionality)
• [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)
• [ ] Infra/Build change
• [ ] Code refactoring

Changes

Checklist:

• [x] I have read and followed the contributing guidelines
• [x] The functionality is complete
• [x] I have commented my code, particularly in hard-to-understand areas
• [x] I have made corresponding changes to the documentation
• [ ] My changes generate no new warnings
• [x] I have added tests that prove my fix is effective or that my feature works
• [x] New and existing unit tests pass locally with my changes

[kunlunl]

OK, I changed it to use warnings instead of raising an error.

[timmoon10]

/te-ci pytorch