tldr Most economists agree that, in a healthy economy, jobs will be both created and destroyed. Typically the destruction is directly tied to creation, in a process known as "Creative Destruction", so naturally there is a preference to emphasise the creation over the destruction. Nevertheless, there may be some special situations where job destruction per se is desirable.
There is some truth in the line of reasoning you seem to be pursuing here, but also some confusion. I'll try to address your various points in turn and hopefully it will clarify the situation. Note that my arguments are based on simple and idealised economic models, which will have many caveats in the real World, but are good enough in the current context.
You are correct to say that, during periods of high unemployment, many economists argue that the government should intervene to create jobs. This is especially true if other economic indicators, such as GDP and inflation, are also weak. The exact way in which the government should do this will depend upon the specifics of the situation, but typically an expansionary fiscal policy is implemented through increased government spending or decreased taxation. This increases aggregate demand which, if the economy is far from full employment, is likely to result in an increase in output without a significant increase in prices. The additional output comes largely from employing the previously unemployed i.e. "creating jobs". However, during periods of full employment (or "labour shortages") such a policy is not generally favoured because the increase in aggregate demand that results is likely to be highly inflationary. Nevertheless, the government will still be keen to "create jobs", but this is a different kind of job creation, as I discuss below.
Suppose the economy is at full employment, this means, to a reasonable approximation, that every worker is doing the job that nets them the greatest possible utility (salary + other effects), in the short term. If you were to "destroy" any of the jobs that people are doing those workers would be worse off, even if many vacancies exist, because they are already in the best job for them. Their employer would also be worse off, because they have lost the output of those workers, which they need to make a profit. The only parties that benefit are the employers with vacancies, but the overall effect is likely to be negative (I'll revisit this last point later). "Destroying" vacant jobs is irrelevant, since nobody is doing them anyway (indeed, they have functionally already been destroyed). An example of this would be setting the minimum wage at a level below the wage of the lowest paid job in the economy. What about creating jobs? This is desirable, provided those jobs offer greater utility to workers than their existing jobs, which is likely to mean that those jobs have higher productivity than at least some of the existing jobs. This sort of job creation is the result of innovation. The overall effect of creating these jobs is that economic output increases, with the benefits being split between workers, employers and the government. There may also be some inflationary effect, because workers have more money to spend, but it is mitigated by the fact that they are also more productive so there are more goods and service to go around.
Notice something about the job creation described in the last paragraph: it also involves job destruction, namely of those jobs that the workers previously performed before moving to their new, higher productivity, jobs. This is one manifestation of the idea of "Creative Destruction" popularised by Joseph Schumpeter, and is a crucial driver of economic growth (at least in the capitalist mode, Marxists have argued that this phenomenon is detrimental in the long run). It is important to recognise the necessity of "destruction" in this process. Indeed, one inhibitor of economic growth is existing institutions (historically usually capital, rather than workers, given its greater power) suppressing innovation in an attempt to prevent destruction of their own enterprises. So you are correct to say that the idea "creation good, destruction bad" is somewhat wrongheaded. Nevertheless, it is natural for people to emphasise creation, because "You'll have a better job!" sounds a lot better than "You'll lose your job!".
In the process I've described above, creation precedes destruction, but it's possible that the causality could be reversed. One of the arguments for introducing a minimum wage is that it will lead to the creation of higher productivity jobs, as employers seek to utilise workers whose productivity would otherwise put them below the minimum wage. However, the evidence for this is mixed and clearly it is somewhat risky to destroy jobs in the hope that better ones will be created. Is there any other circumstance in which the government might want to destroy jobs? One possibility is that the government wishes to move workers into jobs that aren't favoured by the market. Probably the most common instance of this is during a war, when the need for soldiers and weapons exceeds what would be provided by a free market. In this case, it is common for the government to destroy jobs through such measures as taxation and price controls (which decrease supply and set supply below the free market equilibrium, respectively) or more drastically by conscription, which forcibly removes workers from jobs and places them in jobs in the military or military industry. An unusual example is the strategic bombing of German cities during WWII, obviously this wasn't endorsed by the German government, but the American economist JK Galbraith suggested that one of the effects was to deprive workers of jobs in the retail and service sectors and force them into armaments factories, thereby contributing to the German government's aim of increasing military production (although the overall effect remains contentious).
I explained above why, in theory, simply destroying existing jobs is a bad idea. The relationship between unemployment and output (GDP) has also been studied empirically and the result is Okun's Law, which is the observation that a 1% increase in unemployment will cause a 2% decrease in GDP. This is really only an approximate rule, but one does generally find that increasing unemployment leads to lower GDP. It therefore appears that the only politician who would endorse "job destruction" would be one who is only interested in the welfare of the employers who currently can't find workers because their enterprises offer jobs with lower productivity than other existing enterprises. Even if such a politician were to exist, it seems unlikely that they would be open about what they were doing, since it would be harmful to so many different groups!