Contrary to what the QLDoc says, this predicate was way too large to be

evaluated on the 'quick-lint/quick-lint-js' project.

Before:
```
Most expensive predicates for completed query RuleOfTwo.ql:
        time  | evals |   max @ iter | predicate
        ------|-------|--------------|----------
        25m9s |       |              | Declaration#4bfb53be::DirectAccessHolder::thisCouldAccessMember#3#dispred#ffff@8a38e2tm
        17m1s |       |              | Declaration#4bfb53be::DirectAccessHolder::thisCouldAccessMember#3#dispred#fffb@0796c497
         3.5s |   130 | 116ms @ 3    | Declaration#4bfb53be::DirectAccessHolder::thisCanAccessClassTrans#fff@926a68j9
         3.3s |       |              | Declaration#4bfb53be::DirectAccessHolder::thisCouldAccessMember#3#dispred#fffb_1230#join_rhs@25e9ffj8
         1.7s |     3 |  1.7s @ 1    | Element#496c7fc2::ElementBase::toString#0#dispred#ff@fcd81c49
         1.3s |       |              | Declaration#4bfb53be::DirectAccessHolder::thisCouldAccessMember#3#dispred#fffb_0132#join_rhs@9c2065t1
         1.3s |       |              | Declaration#4bfb53be::DirectAccessHolder::thisCouldAccessMember#3#dispred#ffff_0132#join_rhs@672330eh
         1.1s |       |              | Declaration#4bfb53be::DirectAccessHolder::thisCanAccessClassTrans#fff_102#join_rhs@f7d5464o
        829ms |   336 |  85ms @ 6    | Enclosing#c50c5fbf::exprEnclosingElement#1#ff@e34d9wq1
        615ms |       |              | Expr#ef463c5d::Expr::getType#ff@e265e79q
```

After:
```
Most expensive predicates for completed query RuleOfTwo.ql:
        time  | evals |  max @ iter | predicate
        ------|-------|-------------|----------
        11.8s |       |             | _#Class#bacd9b46::Class::getADerivedClass#0#dispredPlus#ff_#Declaration#4bfb53be::AccessHolder::getE__#antijoin_rhs#1@fb0627h8
        4.8s |       |             | _#Class#bacd9b46::Class::getADerivedClass#0#dispredPlus#ff_#Declaration#4bfb53be::AccessHolder::getE__#antijoin_rhs#4@c43dbeia
        3.8s |       |             | _#Class#bacd9b46::Class::getADerivedClass#0#dispredPlus#ff_#Declaration#4bfb53be::AccessHolder::getE__#antijoin_rhs#3313e5963
        3.4s |   130 | 93ms @ 3    | Declaration#4bfb53be::DirectAccessHolder::thisCanAccessClassTrans#fff@a0289bfg
        1.5s |     3 | 1.5s @ 1    | Element#496c7fc2::ElementBase::toString#0#dispred#ff@fcd81c49
        806ms |       |             | Declaration#4bfb53be::DirectAccessHolder::thisCanAccessClassTrans#fff_021#join_rhs@cc1b76s7
        721ms |   336 | 61ms @ 5    | Enclosing#c50c5fbf::exprEnclosingElement#1#ff@e34d9wq1
        489ms |       |             | Expr#ef463c5d::Expr::getType#ff@e265e79q
        337ms |   130 | 62ms @ 5    | Class#bacd9b46::Class::accessOfBaseMemberMulti#ffff@0165b0dr
        329ms |       |             | Variable#7a968d4e::ParameterDeclarationEntry::getAnonymousParameterDescription#0#dispred#ff@0f12bdvq
        211ms |       |             | exprs_10#join_rhs@5481143i
```

Author: MathiasVP

cpp/ql/lib/semmle/code/cpp/Declaration.qll

@@ -619,11 +619,10 @@ private class DirectAccessHolder extends Element {
   /**
    * Like `couldAccessMember` but only contains derivations in which either
    * (5.2), (5.3) or (5.4) must be invoked. In other words, the `this`
-   * parameter is not ignored. This restriction makes it feasible to fully
-   * enumerate this predicate even on large code bases. We check for 11.4 as
-   * part of (5.3), since this further limits the number of tuples produced by
-   * this predicate.
+   * parameter is not ignored. We check for 11.4 as part of (5.3), since
+   * this further limits the number of tuples produced by this predicate.
    */
+  pragma[inline]
   predicate thisCouldAccessMember(Class memberClass, AccessSpecifier memberAccess, Class derived) {
     // Only (5.4) is recursive, and chains of invocations of (5.4) can always
     // be collapsed to one invocation by the transitivity of 11.2/4.