Warn user if bidding in far excess of a smartcoin's global settlement rate

**Is your feature request related to a problem? Please describe.**

If I try to buy 1 bitBTC for 333,333 BTS/BTC, it gives me the following warning:

> Your order is 3.17 times higher than the lowest ask, are you sure?

However, whilst the global settlement rate is only 41,373.60364 BTS , if I try to bid 140,056 BTS (the lowest illiquid ask), I won't be shown a warning that I'm effectively making a similar 300% bid on the value of the backing collateral.

Sure there's subtle indicators that bitBTC is in global settlement, the GS price in the top bar, the inability to borrow new BTC and the ability to immediately settle BTC in the sell box, however some users have expressed dismay at mistakenly bidding way over a rational rate tokens in this manner on the Bitshares DEX before.

**Describe the solution you'd like**
If an asset is in global settlement & the current market trading pair includes the backing collateral asset, then warn the user if they're creating a limit order far in excess of the value of the underlying backing collateral, similar to the current warning if bidding significantly over lowest ask.

**Describe alternatives you've considered**
More clear warnings that the user is bidding on a globally settled asset, rather than checking ratios on order, provide link to global settlement docs.

**Additional context**

[Origin of issue justifying this feature request](https://hive.blog/hive-120117/@nftea.gallery/centralized-lending-platforms-letting-you-down-the-decentralized-credit-offer-tech-on-the-bitshares-blockchain-has-your-back#@felixxx/re-nfteagallery-rm65n7).

No files or tests are named. Start by tracing the existing warning for bids above the lowest ask through the DEX order-entry flow, then inspect how global-settlement rates and backing collateral are exposed; the work is done when an excessive bid on a globally settled asset reliably shows the intended warning and the relevant order-entry checks pass.

Written by the indexing model from the issue text.