Add FlexAttention examples to SDPA tutorial

[dorbittonn]

Summary

• Add new section demonstrating flex_attention from PyTorch 2.5
• Include score_mod examples: relative position bias and ALiBi
• Show block_mask for sparse attention patterns (causal masking)
• Demonstrate combining score_mod and block_mask
• Add performance comparison with standard SDPA

Test plan

• [x] All FlexAttention code tested with PyTorch 2.7.1
• [x] Verified output shapes are correct
• [x] Compilation with torch.compile works as expected

Checklist

• [x] The issue that is being fixed is referred in the description
• [x] Only one issue is addressed in this pull request
• [x] Labels from the issue that this PR is fixing are added to this pull request
• [x] No unnecessary issues are included into this pull request