Adjust release flows to use those with the new branch model (#5015)

This PR contains adjustments of the node release pipelines so that it
will be possible to use those to trigger release actions based on the
`stable` branch.

Previously the whole pipeline of the flows from [creation of the
`rc-tag`](https://github.com/paritytech/polkadot-sdk/blob/master/.github/workflows/release-10_rc-automation.yml)
(v1.15.0-rc1, v1.15.0-rc2, etc) till [the release draft
creation](https://github.com/paritytech/polkadot-sdk/blob/master/.github/workflows/release-30_publish_release_draft.yml)
was triggered on push to the node release branch. As we had the node
release branch and the crates release branch separately, it worked fine.

From now on, as we are switching to the one branch approach, for the
first iteration I would like to keep things simple to see how the new
release process will work with both parts (crates and node) made from
one branch.

Changes made: 

- The first step in the pipeline (rc-tag creation) will be triggered
manually instead of the push to the branch
- The tag version will be set manually from the input instead of to be
taken from the branch name
- Docker image will be additionally tagged as `stable`



Closes: https://github.com/paritytech/release-engineering/issues/214