dmitrygx

@Akshay-Venkatesh could you pls fix code style warnings https://dev.azure.com/ucfconsort/ucx/_build/results?buildId=50138&view=logs&j=cc064a77-22b5-56bf-ecc0-70b5fe764261&t=aedfd754-44a6-53e6-6843-8659d800fee2?

> @Akshay-Venkatesh pls squash I squashed changes and pushed directly to this PR. @yosefe could your review pls?

> @dmitrygx seems some changes were added by force-push, is this expected? > > comparing [1547163](https://github.com/openucx/ucx/commit/15471630a0ecdce01a84a4f74c0bade1863938c7) and [45fac04](https://github.com/openucx/ucx/commit/45fac04aca00dae9c016139b4af152efb0869de7) when merged to [b9000ba](https://github.com/openucx/ucx/commit/b9000bac8e3907f21b441dc705b9d5909767017d) > > * merge of [1547163](https://github.com/openucx/ucx/commit/15471630a0ecdce01a84a4f74c0bade1863938c7) to [b9000ba](https://github.com/openucx/ucx/commit/b9000bac8e3907f21b441dc705b9d5909767017d)...

@yosefe squashed, could you reapprove pls?

@alex--m is it possible to test this PR with the tests introduced in #5055? then we could close #5055 and just create a separate PR with those tests thanks!

@brminich @Artemy-Mellanox could you review pls?

> @dmitrygx seems you forgot to push @brminich yep, fixed

@brminich replaced `UCT_IB_MLX5_MD_TOKEN_SIZE` by `UCT_IB_MLX5DV_FLD_SZ_BYTES`

@yosefe could you review pls?

@brminich could you re-approve pls? changes were squashed and moved the following code to the new place where `cmd_hca_cap` bits are valid (i.e. weren't filled by the next command):