Misc fixes and improvements

What's Changed

• Failing test cases for bit-shifting by @saxbophone in #105
• Optimise multiplication and division by powers of two by @saxbophone in #132
• Rename pow to ipow, change args by @saxbophone in #134

Full Changelog: v0.5.0...v0.5.1

v0.5.0

What's Changed

• Fix divmod bug by @saxbophone in #107
• Josh/67 cout efficiency by @saxbophone in #109
• Changed internal representation to store zero as zero, not empty array by @saxbophone in #112
• Query size of Nat object by @saxbophone in #113
• Josh/117 decrement 1 bug by @saxbophone in #118
• Attempting to add an additional github-actions build that also runs t… by @saxbophone in #106
• Initial reimplementation of storage type selection: by @saxbophone in #116
• Initial implementation of digits -> Nat -> digits by @saxbophone in #120

Full Changelog: v0.4.2...v0.5.0

v0.4.2

What's Changed

• Docs update v0.4 + efficiency increases by @saxbophone in #101

Full Changelog: v0.4.1...v0.4.2

Minor docs fix

v0.4.1

Minor docs fix

v0.4.0

What's Changed

• Import literals namespace into parent namespace by @saxbophone in #94
• Move pow to math support library and added ilog by @saxbophone in #95
• Implement bitwise operators by @saxbophone in #96
• 47 other cast operators by @saxbophone in #74

Full Changelog: v0.3.1...v0.4.0

Bugfix release v0.3.1

• Nat::from_float() no longer produces leading zeroes and error when called with float in range 0..<1
• Changed return type of Nat::divmod() from tuple to pair
• Precompile arby's header when used in the test suite
• Misc. docs changes

---

Here follows an auto-generated summary of the changes:

What's Changed

• feat divmod: change return type to std::pair by @deadb0d4 in #87
• Bugfix: leading zero error when creating Nat from float by @saxbophone in #92

New Contributors

• @deadb0d4 made their first contribution in #87

Full Changelog: v0.3.0...v0.3.1

Open-sourced v0.3.0

• renamed Uint to Nat
• switched digits storage to custom constexpr doubly-linked-list to support constexpr on platforms lacking constexpr std::vector
• added more unit tests and fixed some bugs

Now almost entirely constexpr!

Using a custom-made constexpr doubly-linked-list type, Uint is now almost entirely constexpr, except for conversion to and from string, and for conversion from float.

Basic version

No bit-shifting in this one, unsigned only.