Solve emits too many errors

[HannesWell]

Describe the bug

When solving equations often too many or unnecessary error messages are printed, even when the equation(-set) was finally solved successfully.

To Reproduce and expected behavior

>> Solve({8.0E-9==x0/x1,x0==x1^4.0},{x0,x1}) 
Error: 
  Power: Infinite expression 1/0 encountered. 
  Infinity: Indeterminate expression 0*ComplexInfinity encountered.
{{x0−>29.77443,x1−>2.33594},{x0−>-14.88721−25.78541⋅ⅈ,x1−>-1.16797−2.02298⋅ⅈ},{x0−>-14.88721+25.78541⋅ⅈ,x1−>-1.16797+2.02298⋅ⅈ}}

In this case I would expect to get just the solution without an error message, since the solution was computed successfully. As we have already discussed the different attempts to find a solution should probably done in 'quite'-mode?

>> Solve({Sin(x0)==5.0},{x0})
Error: Using inverse functions may omit some solutions.
{{x0−>1.5708−2.29243⋅ⅈ}}

I think the message about potentially omitted solutions is at most a warning and maybe should only be displayed in verbose mode (does something like that exist in symja?), by default it should be maybe an INFO Level message.

>> Solve({x0^2.0*Sin(x1)==5.0,x1^3.0*Cos(x0)==5.0},{x0,x1})
Error: 
  FindRoot: interval does not bracket a root: f(NaN) = NaN, f(NaN) = NaN. 
  Solve: This system cannot be solved with the methods available to Solve
Solve({x02.0⋅Sin(x1)==5.0,x13.0⋅Cos(x0)==5.0},{x0,x1})

Although the given set of equations cannot be solved, I think the error message FindRoot: interval does not bracket a root: f(NaN) = NaN, f(NaN) = NaN. should not be emitted. This has likely the same cause as the first example.

Java

• Version: Java 11
• OS: all

[HannesWell]

The FindRoot warnings for the last case are suppressed thanks to your last commit. But in the first case the warnings are still printed to the console.