at the time of submitting samples for NMR or HRMS, we should confirm the solubility of compounds. Sometimes a compound is soluble in just adding the solvent and sometimes using sonication, it creates confusion during the solubility test.
in my opinion, Sonication is not essential for testing solubility. But if the sample is not easily soluble in a particular soluble (say partially soluble in CDCl3) then you can use sonication to enhance the solubility because sonication leads to acoustic cavitation which generates bubbles having high pressure and temperature and this helps in mixing of two or more substances. but if your compound is easily soluble in a solvent then there is need of sonication at all
Sonication is not inevitable in making a solute to dissolve in a given solvent. Also sonication is not considered as a test of solubility though the ultrasonic parameters measured differ if the solute is not completely soluble as the parameters depend on the concentration of the solution. Sonication definitely enhances the solubility or helps solubility by i) decreasing the particle size ii) a sort of effective stirring.
For the sake of making samples for NMR, what is needed is to ascertain whether the required amount of the sample is soluble in the required volume of the solvent or not. For this you may not need sonication. Please see the reasons behind sonication, as mentioned by Kanakam.
In a standard solubility test you determine the maximum amount of a substance that is soluble in a particular volume of a solvent at a particular temperature. Here, you are making a saturated solution and sonication may be required to facilitate dissolution.