[ConstraintSystem] Improve score label in the debug output so that we can easily recognize which score is more/less impactful.

[kntkymt]

Motivation

Environment

swift-driver version: 1.90.11.1 Apple Swift version 5.10 (swiftlang-5.10.0.13 clang-1500.3.9.4)
Target: arm64-apple-macosx14.0

In current score label in the debug output,
We cannot easily recognize which score is more impactful.

class Animal {}
class Cat: Animal {}
func f(_ a: [Cat?]) { }
func f(_ a: [Any]) { }
func f(_ a: [Animal]) {}
let a: [Cat] = [Cat()]

f(a)

// swiftc -Xfrontend -debug-constraints Source.swift

---Solver statistics---
--- Solution #0 ---
Fixed score: [component: collection upcast conversion(s), value: 1] [component: value to optional promotion(s), value: 1]
...

--- Solution #1 ---
Fixed score: [component: collection upcast conversion(s), value: 1] [component: empty-existential conversion(s), value: 1]
...

--- Solution #2 ---
Fixed score: [component: value-to-pointer conversion(s), value: 1]
...

---Solution---
Fixed score: [component: value-to-pointer conversion(s), value: 1]
...

Full Output

We cannot recognize which is more/less impactful between value to optional promotion of Solution#0 and empty-existential conversion of Solution#1.

In Addition, Since Solution is value-to-pointer conversion(s), We can guess value-to-pointer conversion is least impactful, but it is also ambiguous.

Proposed solution

Add the number indicates impactful in the debug output. We can recognize which score is more/less impactful comparing the number.

---Solver statistics---
--- Solution #0 ---
Fixed score: [component: #13 collection upcast conversion(s), value: 1] [component: #14 value to optional promotion(s), value: 1]
...

--- Solution #1 ---
Fixed score: [component: #13 collection upcast conversion(s), value: 1] [component: #15 empty-existential conversion(s), value: 1]
...

--- Solution #2 ---
Fixed score: [component: #17 value-to-pointer conversion(s), value: 1]
...

---Solution---
Fixed score: [component: #17 value-to-pointer conversion(s), value: 1]
...

I think these format is also good.

// as suffix?
[component: value-to-pointer conversion(s) #17, value: 1]

// print reversed index value instead of enum value?
[component: #3 value-to-pointer conversion(s), value: 1]

I think It will be easy to implement this feature.

https://github.com/hborla/swift/blob/6676fbfd19781dd6ed1d59bb84146aff336c24d4/include/swift/Sema/ConstraintSystem.h#L1152

  void print(llvm::raw_ostream &out) const {
    bool hasNonDefault = false;
    for (unsigned int i = 0; i < NumScoreKinds; ++i) {
      if (Data[i] != 0) {
        out << " [component: ";
        out << "#";
        out << std::to_string(i);
        out << " ";
        out << getNameFor(ScoreKind(i));
        out << "(s), value: ";
        out << std::to_string(Data[i]);
        out << "]";
        hasNonDefault = true;
      }
    }
    if (!hasNonDefault) {
      out << " <default ";
      out << *this;
      out << ">";
    }
  }

Alternatives considered

Restore the score format used before these PRs were implemented.

#60387
#60927

---Solver statistics---
--- Solution #0 ---
Fixed score: 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0
...

--- Solution #1 ---
Fixed score: 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0
...

--- Solution #2 ---
Fixed score: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
...

---Solution---
Fixed score: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
...

However, I think it is better to show name of component. So I propose mixed format showing name and number.

Additional information

If it is acceptable, I'd like to challenge creating PR by myself 😎 (It's my first time to contribute).