harden: make the background physically unable to read the shared state singleton (closes #324)

Five defects traced to one fact: src/background/index.js read and wrote the
module-level `state` singleton in src/shared/state.js, which the MV3 service
worker never populates and which answered an unpopulated read out of
DEFAULT_STATE in silence. Every previous fix added a loadState() before the
access, and that is what produced the fifth: a load detaches the objects an
in-flight handler is holding.

So the reachability goes rather than a sixth call site.

The background now has its own storage layer, src/background/state.js:
getState() is a detached, normalized per-call read, and updateState() is a
queued read-modify-write whose read is one storage round trip ahead of its
write. Nothing in the background holds an in-memory copy of the profile.

- Every handler takes one snapshot and answers from it, including the address
  it names: activeAddressOf(s) replaced a second, later storage read that
  could disagree with the first.
- wallet_switchEthereumChain applies applyChainSwitchFields() (split out of
  chainSwitch.js, which keeps the singleton path for the popup) inside
  updateState() instead of calling onChainSwitch() on the singleton.
- The remembered site decision is a read-modify-write, not a load-mutate-save
  around a prompt the user takes seconds to answer.
- backgroundRefresh() refreshes a private copy of the wallets and applies the
  balances that came back by address, so it never publishes an object other
  in-flight work holds, and a wallet added or deleted during the round trip
  survives its write.
- The transaction attempt takes its chain id and its endpoint from the same
  snapshot. They used to come from different moments, so a chain switch
  committed in between moved the endpoint under an artifact already verified
  against the old chain.

getProvider(rpcUrl, networkId) now REQUIRES the network id and validates it
against networks.js. That closes the cold-worker wrong-chain send at its shape
rather than at one call site: the hint used to default to currentNetwork() off
the unpopulated singleton, so the endpoint was the user's chain and ethers
fixed chainId at 0x1, and the wallet's own verifySignedTx then refused every
non-mainnet dApp send. refreshBalances(), lookupTokenInfo(), scanForAddresses()
and resolveEnsName() carry the id through; balances.js no longer requires
state.js at all.

The prohibition is enforced mechanically, not by review, and it is enforced by
the bundler rather than by a guess at what the bundler does. build.js keeps a
FORBIDDEN_INPUTS table of modules an entry point's bundle may not contain, and
assertNoForbiddenInputs() fails the build when esbuild's metafile reports
src/shared/state.js as an input of a background bundle, naming the import chain
from the metafile's own graph. That is the resolution the shipped bundle was
built from, so no specifier syntax, no hop and no resolution rule can slip past
it; Dockerfile:42 runs make build, so it holds in CI. A FORBIDDEN_INPUTS key
that matches no bundled entry point also fails, so the table cannot rot into a
vacuous pass.

A custom ESLint rule walks the CommonJS require graph from every src/background/
file and reports the same thing in the editor, before a full bundle. It matches
specifiers textually, so it is best-effort fast feedback and not the guarantee —
two earlier revisions of it shipped holes (a template literal, a dynamic
import(), a comment inside the call, a directory resolved through package.json
main). Those are covered now and pinned by
tests/backgroundStateLintRule.test.js, and the next divergence between a
hand-rolled matcher and a real bundler is caught by the build instead. A
computed specifier (require("../shared/" + "state")) is deliberately not
matched: esbuild cannot resolve it either, so it never reaches the bundle.

Reading a persisted field of the singleton before any load now throws
StateNotLoadedError instead of serving DEFAULT_STATE.

Test stubs: chrome.storage.local is a serialization boundary, and eight files
stubbed it with an aliasing get, so the object a module held and the object
"storage" held were one object — an assertion could pass on a build that never
wrote anything. Every test that drives real persistence now goes through
tests/support/storageStub.js, which structured-clones in both directions.

closes #320

TODO.md

@@ -45,6 +45,37 @@ but the review is broader than any of them.
 
 # Completed Steps
 
+- 2026-08-23: The background no longer reads or writes the shared `state`
+  singleton ([#324](https://git.eeqj.de/sneak/AutistMask/issues/324)), which
+  also closes the cold-worker wrong-chain send
+  ([#320](https://git.eeqj.de/sneak/AutistMask/issues/320)). One in-memory copy
+  loaded once is the popup's lifetime, not the MV3 worker's: the worker is
+  killed when idle, nothing loaded state at module scope, and an unpopulated
+  read was answered out of `DEFAULT_STATE` in silence. Five defects traced to
+  that, and every point fix added a `loadState()` that created the next one — a
+  load detaches the objects an in-flight handler is holding. The background now
+  has its own storage layer (`src/background/state.js`): `getState()` for a
+  detached per-call read, `updateState()` for a queued read-modify-write.
+  `backgroundRefresh()` refreshes a private copy and applies the balances that
+  came back by address, so a wallet added, renamed or deleted during the round
+  trip survives. The transaction attempt takes its chain id and its endpoint
+  from one snapshot, so a committed chain switch can no longer move the endpoint
+  under an artifact already verified against the old chain. `getProvider()` now
+  REQUIRES the network id, which is what closes
+  [#320](https://git.eeqj.de/sneak/AutistMask/issues/320) at the shape rather
+  than at the call site. The prohibition is enforced by `build.js`, which fails
+  the build when esbuild's own metafile reports `src/shared/state.js` as an
+  input of either background bundle — the resolution the shipped bundle was
+  actually built from, so no specifier syntax and no resolution rule can slip
+  past it, and `make build` runs in CI. An ESLint rule that walks the require
+  graph textually gives the same answer in the editor, before a full bundle; it
+  is fast feedback rather than the guarantee, and the shapes it is known to
+  catch are pinned by `tests/backgroundStateLintRule.test.js`. Reading an
+  unloaded singleton now throws `StateNotLoadedError` instead of serving
+  defaults. The `chrome.storage.local` stubs in eight test files aliased instead
+  of structured-cloning, which could let an assertion pass on a build that never
+  wrote anything; every test that drives real persistence now goes through
+  `tests/support/storageStub.js`.
 - 2026-08-23: A swap always names its output token
   ([#346](https://git.eeqj.de/sneak/AutistMask/issues/346)). The `Token Out`
   detail line in `src/shared/uniswap.js` was pushed only when a symbol was