Remove redundant default constructors when they're the only constructor (#8461)

* Initial plan

* Modernize C++ constructors to use C++11 default member initialization - Phase 1

Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

* Fix theory_pb.h struct definition - move reset() back inside struct

Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

* Modernize C++ constructors to use C++11 default member initialization - Phase 2

Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

* Fix opt_solver.h - revert rational initialization (complex type)

Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

* Modernize C++ constructors to use C++11 default member initialization - Phase 3

Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

* Fix sparse_matrix.h - explicitly initialize union member in default constructor

Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

* Remove unnecessary default constructors when they're the only constructor

Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: nunoplopes <2998477+nunoplopes@users.noreply.github.com>

src/util/approx_nat.h

@@ -23,10 +23,10 @@ Notes:
 #include<climits>
 
 class approx_nat {
-    unsigned m_value;
+    unsigned m_value = 0;
     static const unsigned m_limit = UINT_MAX >> 2;
 public:
-    approx_nat():m_value(0) {}
+    approx_nat() = default;
     explicit approx_nat(unsigned val);
     bool is_huge() const { return m_value == UINT_MAX; }
     unsigned get_value() const { return m_value; }

src/util/checked_int64.h

@@ -34,14 +34,14 @@ class overflow_exception : public z3_exception {
 
 template<bool CHECK>
 class checked_int64 {
-    int64_t m_value;
+    int64_t m_value = 0;
     typedef checked_int64 ci;
     
     rational r64(int64_t i) const { return rational(i, rational::i64()); }
 
 public:
 
-    checked_int64(): m_value(0) {}
+    checked_int64() = default;
     checked_int64(int64_t v): m_value(v) {}
 
     bool is_zero() const { return m_value == 0; }

src/util/ema.h

@@ -21,11 +21,11 @@ Revision History:
 #pragma once
 
 class ema {
-    double m_alpha, m_beta, m_value;
-    unsigned m_period, m_wait;
+    double m_alpha = 0, m_beta = 1, m_value = 0;
+    unsigned m_period = 0, m_wait = 0;
     bool invariant() const { return 0 <= m_alpha && m_alpha <= m_beta && m_beta <= 1 && m_wait <= m_period; }
  public:
-    ema(): m_alpha(0), m_beta(1), m_value(0), m_period(0), m_wait(0) {
+    ema() {
         SASSERT(invariant());
     }
 

src/util/inf_s_integer.h

@@ -25,8 +25,8 @@ class inf_s_integer {
     static inf_s_integer m_zero;
     static inf_s_integer m_one;
     static inf_s_integer m_minus_one;
-    int m_first;
-    int m_second;
+    int m_first = 0;
+    int m_second = 0;
  public:
 
     unsigned hash() const { 
@@ -44,7 +44,7 @@ class inf_s_integer {
 
     std::string to_string() const;
 
-    inf_s_integer():m_first(0), m_second(0) {}
+    inf_s_integer() = default;
 
     explicit inf_s_integer(int n):m_first(n), m_second(0) {}
     explicit inf_s_integer(int n, int d): m_first(n), m_second(0) { SASSERT(d == 1); }

src/util/mpbq.h

@@ -31,11 +31,11 @@ Revision History:
 #include "util/vector.h"
 
 class mpbq {
-    mpz      m_num;
-    unsigned m_k;    // we don't need mpz here. 2^(2^32-1) is a huge number, we will not even be able to convert the mpbq into an mpq
+    mpz      m_num = 0;
+    unsigned m_k = 0;    // we don't need mpz here. 2^(2^32-1) is a huge number, we will not even be able to convert the mpbq into an mpq
     friend class mpbq_manager;
 public:
-    mpbq():m_num(0), m_k(0) {}
+    mpbq() = default;
     mpbq(int v):m_num(v), m_k(0) {}
     mpbq(int v, unsigned k):m_num(v), m_k(k) {}
     mpz const & numerator() const { return m_num; }