Improve performance in Fill addition

[jishnub]

On master

julia> F = Fill(SMatrix{2,2}(1:4), 2, 2); M = Matrix(F);

julia> @btime $F + $M;
  65.964 ns (2 allocations: 288 bytes)

This PR

julia> @btime $F + $M;
  44.739 ns (1 allocation: 192 bytes)

We avoid the extra allocation of a 1-element vector that is there to only ensure scalar-like broadcasting. The same may be achieved by using a Tuple.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.90%. Comparing base (4bfbeec) to head (b68c55c).
Report is 7 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #371   +/-   ##
=======================================
  Coverage   99.90%   99.90%           
=======================================
  Files           8        8           
  Lines        1086     1106   +20     
=======================================
+ Hits         1085     1105   +20     
  Misses          1        1           

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