I wouldn't say that is mistreatment. It is a business transaction where one party has more leverage than the other. 99% of the time the employee in the lower income region gets a far high salary than they expect to get at another employer. So it is not like they are being coerced into a "unable to make ends meet" levels of income.
Realistically, the upper limit of a salary is usually determined by the average value the employee will bring to the company because profit is derived from surplus value. This value may range from very direct (number of product produced per unit of time) or very indirect (making the rest of the team happy to increase their overall productivity and reduce employee turnover).
The lower limit is generally whatever the company can get away with. This has legal factors (e.g. minimum wage laws) as well as regional (e.g. "market rate", unemployment rates) but ultimately the candidate has to make the call what they can (or have no other choice but to) accept.
Crucially, the surplus value only hinges on the value generated and the expenses of employing that person (mostly salary and benefits but cross-state and internationally there may be legal costs avoidable by hiring locally).
If you live in an area with a low cost of living and the company's upper limit is defined by the relative cost of living, they're going to extract a lot more surplus value from your labor than if you live in an area with a high cost of living but maintain the same personal cost of living because you manage to be extraordinarily frugal.
That you make a lot of money relative to your local peers does not mean the company isn't disproportionately extracting surplus value from you (or in other words: paying you in lower proportion to the value you generate than others). If everyone else in your neighborhood is paid a penny and you earn a dime that does not mean I'm not exploiting you when I pay the locals a dollar for the same generated value.
That GitLab only varies pay and not the cost of their software and services demonstrates that they understand this. The value their services provide depends entirely on the customer's market (which is btw why a lot of initial SaaS pricing is based on US startup culture where company expenses are measured as burn rates rather than profitability, even if it makes the services cost prohibitive in other markets).
Also of course geographically distributed employees will find it harder to unionize (even more so formally because of different jurisdictions and labor laws) or organize and are less likely to seriously develop solidarity in a way that could get in the way of making a profit.
There are arguments for the destructive impact of foreign companies paying above market rate but it's absurd to think companies that use geographical salary ranges do so out of concern for regional competitors.
Sure, and I'm talking about how much a company is willing to pay. A company acting economically is not going to pay more for a developer than the value they will provide, or else hiring them will lose them money.
Note that as I said, value can be much less straightforward to calculate. An employee may provide no direct value but have a positive effect on the other employees, increasing their value generation. Or an employee may provide excellent value but drive down everyone else's productivity (e.g. by making them unhappy or constantly demanding attention) thus reducing overall value. Or an employee may initially provide low value but have massive potential.
Additionally the problem with hiring is that the candidate has yet to generate any value at all for the company and even if they do have massive potential there is no guarantee that the company can tap into that potential fully. So we're not even talking value (as when considering a pay rise) but expected value.
The cost of labor is caused by paying employees wages/expenses/benefits and having administrative and legal overhead for doing that. The wages and benefits are to a degree subject to demand and, to an even lesser degree, supply.
What a candidate is willing to accept on the other hand depends on what they think they could get elsewhere, how much they want to work for the company in question in particular, what their personal expenses are and how badly they need money.
Companies generate profit by extracting surplus value by simply paying employees less than the value they generate (companies don't produce value, labor does, profit is surplus value left in the company by paying labor less than the value it generates, adjusted for the cost of doing business).
Geographical salaries work by placing the company in one location (usually with a relatively high cost of living) and hiring employees in various other locations (usually with a relatively lower cost of living).
If you can get people from poorer countries producing the same amount of value for your company as people from your richer country, and you can use geographical salaries to justify paying them less, you increase your profit margin on that labor simply by replacing a highly paid local position with a less highly paid remote position.
In the 90s we used to call this "outsourcing" except we did it with entire teams of people rather than individuals (to mixed success because it hinged on the team lead not only understanding you correctly but also propagating the information correctly and also usually reduced overall visibility outside the team). It was widely understood as the blatant cost-cutting measure it was.
Calling it "geographical salaries" and doing it at a finer resolution doesn't change its nature.
And yeah, you won't get a fairer share of your value just by asking for it individually. You'll just get fired and replaced by someone who's happy with what they're offered. This is what you need collective bargaining for -- except that's much harder to organize or even decide on when you're geographically distributed and mostly using company controlled channels for communication.
The employee ends up doing pretty well.