How does repairing their gear not qualify as a casus belli for Russia?

This is actually an interesting question.

Traditionally, doing things like supplying arms to Ukraine would have been treated as making states party to the conflict.

The standards for what is expected of a neutral party have kind of shifted over time. At one point, there was an expectation that a neutral party could not treat parties to a conflict differently; if they would permit arms shipments to one side, they must also permit arms shipments to another.

Over time, various parties have kind of decided that they are still going to distinguish between parties directly involved in a conflict and parties aiding one side.

https://lieber.westpoint.edu/future-law-of-neutrality/

Neutral Obligations under the Traditional Law of Neutrality

The traditional law of neutrality was developed to enable non-belligerent States to maintain peaceful relations with belligerent parties involved in armed conflict and continue trading without participating in hostilities. As such, neutral States have an obligation to abstain from providing war-related goods and services to belligerent parties (§15.3.2.1). However, this neutral obligation was qualified in three respects when these rules were codified in 1907 under Hague Convention V regarding neutrality in land warfare and Hague Convention XIII regarding neutrality in naval warfare.

First, the obligation of abstention is a specific one, applying to the provision of “munitions of war or supplies” (Article 2 and Article 5 of Hague Convention V) and “warships, ammunition, or war material” (Article 6 of Hague Convention XIII) or associated services such as forming corps of combatants on its territory (Article 4 of Hague Convention V) and arming vessels within its jurisdiction (Article 8 of Hague Convention XIII). This means that under general international law other types of support, such as financial assistance or the provision of goods and services that are unrelated to the conduct of hostilities, fall outside of neutral obligations (cf. Article 16(2) of Havana Convention on Maritime Neutrality). For example, Switzerland has provided humanitarian support to the victims of the war in Ukraine, while Japan pledged a total of $600 million in financial assistance.

The imposition of unilateral sanctions against Russia, as well as weapons export and intelligence sharing in support of Ukraine, may appear to run contrary to these neutral obligations. So, what does this recent practice mean for the future development of neutrality law?

Weapons Export

The government authorization for the export of weapons and military equipment is a clear example of unneutral support. When the Lend-Lease Act was enacted in 1941 to facilitate arms supply to Great Britain and other countries, it triggered concerns that it would permit departure from neutral obligations under international law. As Quincy Wright observed at that time, the Act constituted “the first legislative endorsement since the Napoleonic period” of the abandonment of impartiality in the interest of American defense (p. 313). Eighty-one years later, the U.S. Congress passed the Ukraine Democracy Defense Lend-Lease Act of 2022, authorizing the Biden administration to lend or lease military equipment to Ukraine and other Eastern European countries impacted by Russia’s invasion of Ukraine.

The official endorsement of arms export in support of Ukraine indicates a shift in State practice relevant to the law of neutrality. This shift is taking place due to the geopolitical situation surrounding the Russia-Ukraine war. An important geopolitical factor is that Russia, as the disadvantaged belligerent due to this unneutral assistance, is not in a position to enforce its belligerent rights against neutral States. There are various means to enforce belligerent rights against unneutral assistance, such as capture of neutral merchant vessels carrying contraband. However, Russia has not issued contraband lists to restrict shipments to Ukraine, presumably because it has characterized the engagement in Ukraine as a “special military operation.” As an alternative, the requirement may be satisfied by a listing of exempt goods (§7.4.1), yet, Russia has been blocking neutral trade entirely.

It comes as no surprise that some States are inclined to lend support to a politically, economically or ideologically favored belligerent against another. Indeed, the past practices suggest that States have tried to circumvent or abdicate neutral obligations based on the doctrine of “non-belligerency” or “qualified neutrality.” As Heintschel von Heinegg observed, such practices on the part of some neutral States deviating from the strict rules of neutrality alone are not sufficient as evidence of a corresponding change in the law of neutrality (pp. 553-6). However, Russia’s unwillingness to take remedial measures (other than targeting imported weapons or jamming satellites in the conduct of hostilities), even when there are means at its disposal to enforce belligerent rights in accordance with the traditional law of neutrality, could be indicative of a change in legal considerations.

The current legal landscape is indeed significantly different to the one in which the traditional law of neutrality was developed. As I discussed elsewhere, contemporary international law allows neutral States to invoke the right of collective self-defense as a justification for violations of neutral obligations by supporting one of the belligerent parties (pp. 133-4). This means that even without being part of NATO, Ukraine has the inherent right to request military intervention by other States to help defend themselves, and other States are entitled to provide support in response to such a request. Self-defense justifies not only a use of force which is otherwise prohibited under customary international law but also non-performance of obligations associated with it as a “circumstance precluding wrongfulness” under the law of State responsibility (Article 21 of Articles on State Responsibility). As Michael N. Schmitt has explained, a State would not necessarily lose neutral status by providing unneutral support in the exercise of the right of collective self-defense.

The shift toward “benevolent neutrality” appears to be further facilitated due to practical difficulties in controlling or challenging the provision of support. When the traditional law of neutrality was developed, the primary means of transport was shipping and belligerent parties deployed naval forces to stem the flow of arms trade through maritime routes. In modern days, by contrast, the supply of weapons and military equipment can be airlifted or, with the use of 3D printing, can be locally manufactured through the transmission of necessary data. This trend is likely to continue, with a greater investment in and use of smaller, scalable, and maneuverable unmanned platforms which have proven to be effective against conventional instruments of war. The Russia-Ukraine conflict has demonstrated that neutral States would find little constraint in providing unneutral support when there is no feasible means to detect and intercept such transfer.

The article also talks about the implications of intelligence sharing on neutrality.

The US has taken the position that providing intelligence to a party in a conflict – as the article points out, which is being done with Ukraine – is okay for a non-belligerent. That is not something that everyone has historically decided they agreed with. In World War II, US warships – theoretically still neutral at that point – would look for German U-Boats and, while not attacking them, would tell the British where they were, who then attacked them.

In World War II, the US considered herself neutral during the Greer incident:

https://en.wikipedia.org/wiki/USS_Greer

At 0840 that morning, Greer, carrying mail and passengers to Iceland, “was informed by a British plane of the presence of a submerged submarine about 10 miles [(16 km)] directly ahead. … Acting on the information from the British plane the Greer proceeded to search for the submarine and at 0920 she located the submarine directly ahead by her underwater sound equipment. The Greer proceeded then to trail the submarine and broadcast the submarine’s position. This action, taken by the Greer, was in accordance with her orders, that is, to give out information but not to attack.” The British plane continued in the vicinity of the submarine until 1032, but prior to her departure the plane dropped four depth charges in the vicinity of the submarine. The Greer maintained [its] contact until about 1248. During this period (three hours 28 minutes),the Greer manoeuvred so as to keep the submarine ahead. At 1240 the submarine changed course and closed the Greer. At 1245 an impulse bubble (indicating the discharge of a torpedo by the submarine) was sighted close aboard the Greer. At 1249 a torpedo track was sighted crossing the wake of the ship from starboard to port, distant about 100 yards [(100 m)] astern. At this time the Greer lost sound contact with the submarine. At 1300 the Greer started searching for the submarine and at 1512 … the Greer made underwater contact with a submarine. The Greer attacked immediately with depth charges.[6]