add istree for Nums

[isaacsas]

For https://discourse.julialang.org/t/symbolics-istree-returning-false-when-there-is-a-tree/82993/3

[codecov-commenter]

Codecov Report

Merging #625 (5707211) into master (9c5deaf) will decrease coverage by 74.70%.
The diff coverage is 0.00%.

@@            Coverage Diff             @@
##           master    #625       +/-   ##
==========================================
- Coverage   77.95%   3.25%   -74.71%     
==========================================
  Files          23      23               
  Lines        2377    2306       -71     
==========================================
- Hits         1853      75     -1778     
- Misses        524    2231     +1707     

Impacted Files	Coverage Δ
src/num.jl	0.00% <0.00%> (-67.82%)	⬇️
src/groebner_basis.jl	0.00% <0.00%> (-100.00%)	⬇️
src/semipoly.jl	0.00% <0.00%> (-95.37%)	⬇️
src/linear_algebra.jl	0.00% <0.00%> (-89.48%)	⬇️
src/arrays.jl	0.00% <0.00%> (-87.07%)	⬇️
src/array-lib.jl	0.00% <0.00%> (-86.51%)	⬇️
src/diff.jl	0.00% <0.00%> (-79.86%)	⬇️
src/wrapper-types.jl	0.00% <0.00%> (-78.75%)	⬇️
src/complex.jl	0.00% <0.00%> (-77.28%)	⬇️
src/build_function.jl	0.00% <0.00%> (-73.51%)	⬇️
... and 12 more

📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more

[isaacsas]

I would add that if this isn't considered appropriate I think an error message should be given indicating that one needs to unwrap the Num. Otherwise one can just get a wrong answer due to https://github.com/JuliaSymbolics/TermInterface.jl/blob/91f77ff2babaaad22a6a2a0c6adbb6a2b7ccb149/src/TermInterface.jl#L3-L9

[shashi]

Num is intentionally not treated as a term, so that we can keep some kind of modularity between dispatch as Real vs expr types...