Fix inferrability of size(::AdjTrans)

[timholy]

For tuples, the compiler can't infer the size of t[[2; 1]]:

julia> revsize1(A::AbstractMatrix) = size(A)[[2; 1]]
revsize1 (generic function with 1 method)

julia> revsize2(A::AbstractMatrix) = reverse(size(A))
revsize2 (generic function with 1 method)

julia> A = rand(3, 5);

julia> @code_warntype revsize1(A)
MethodInstance for revsize1(::Matrix{Float64})
  from revsize1(A::AbstractMatrix) @ Main REPL[1]:1
Arguments
  #self#::Core.Const(revsize1)
  A::Matrix{Float64}
Body::Tuple{Vararg{Int64}}
1 ─ %1 = Main.size(A)::Tuple{Int64, Int64}
│   %2 = Base.vcat(2, 1)::Vector{Int64}
│   %3 = Base.getindex(%1, %2)::Tuple{Vararg{Int64}}
└──      return %3


julia> @code_warntype revsize2(A)
MethodInstance for revsize2(::Matrix{Float64})
  from revsize2(A::AbstractMatrix) @ Main REPL[2]:1
Arguments
  #self#::Core.Const(revsize2)
  A::Matrix{Float64}
Body::Tuple{Int64, Int64}
1 ─ %1 = Main.size(A)::Tuple{Int64, Int64}
│   %2 = Main.reverse(%1)::Tuple{Int64, Int64}
└──      return %2

[Copilot]

Pull request overview

This pull request fixes a type inference issue in the size method for adjoint and transpose linear operators by replacing tuple indexing with the reverse function.

Changes:

• Replace size(A)[[2; 1]] with reverse(size(A)) for AdjTrans operators to enable better type inference by the Julia compiler

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.44%. Comparing base (32dbc5e) to head (0174e69).
⚠️ Report is 68 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #403      +/-   ##
==========================================
+ Coverage   95.00%   95.44%   +0.44%     
==========================================
  Files          17       21       +4     
  Lines        1100     1186      +86     
==========================================
+ Hits         1045     1132      +87     
+ Misses         55       54       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.