Lux: Explicit Parameterization of Deep Neural Networks in Julia

LuxLib LuxLib-v1.7.0

Diff since LuxLib-v1.6.2

Merged pull requests:

• docs: update readme (#1265) (@avik-pal)
• feat: add MultiHeadAttention (#1266) (@avik-pal)
• docs: use documenter citations (#1267) (@avik-pal)
• refactor: revert runic formatting (#1268) (@avik-pal)
• chore: bump crate-ci/typos from 1.30.1 to 1.30.2 (#1270) (@dependabot[bot])
• chore: remove unused imports (#1271) (@avik-pal)
• fix: reactant typo (#1275) (@avik-pal)
• fix: outputsize of Dense (#1276) (@avik-pal)

Closed issues:

• Emit Batchnorm Op for Training (#1208)
• outputsize(::Dense) (#1262)