MLIR - Adding Max and Square ReLU operations to the Toy dialect and Lowering to LLVM

YOU CAN REPLICATE THIS PROJECT -> https://github.com/Nagharjun17/MLIRX

Added custom MLIR dialect ops SquareReLU and Max to the Toy Ch6 example (ODS, builders, Dialect integration).
Implemented lowering: MaxOp → Affine loops (CmpF, Select) and SquareReLU → Mul + Max.
Extended MLIRGen + CMake to support new ops and regenerate code.
Validated full pipeline by building with Ninja and emitting LLVM IR via toyc-ch6 -emit=llvm.